The Prenatal Origin of Behavior by Davenport Hooker, PH.D., Sc.D. - University of Kansas Press, Lawrence, Kansas, 1952

Tne Prenatal Origin
of Behavior
The Prenatal Origin
of Beliavior
t y
Davenport Hooker, Pli.D., Sc.D.
Professor of Anatomy and Chairman of the Department
University of Pittsburgh School of Medicine
Porter Lectures, Series 18
University of Kansas Press, Lawrence, Kansas, 1952
COPYRIGHT, 1952, BY
THE UNIVERSITY OF KANSAS PRESS
Physiological and morphological studies on human prenatal de-
velopment, publication no. 20. These studies have been aided by-
grants from the Penrose Fund of the American Philosophical So-
ciety, from the Carnegie Corporation of New York, from the
University of Pittsburgh, and from the Sarah Mellon
Scaiie Foundation.
Preface
T
o BE INVITED to fill the Porter Lectureship in Medi-
cine is indeed an honor and one for which this lec-
turer is most grateful. Quite aside from my personal
gratification at being invited to lecture on this foundation,
I am especially pleased because of my long friendship with
Dr. George Ellett Coghill, the founder in America of work
on embryonic movements, who served the University of
Kansas School of Medicine from 1913 to 1925. It has also
been my privilege to know for a long time both Dr. Henry
Carroll Tracy, his successor as Chairman of the Depart-
ment at Kansas, and the present Chairman, Dr. Paul Gib-
bons Roofe.
I am indebted for many things to many people too
numerous to name here, but it would be falling short of
both justice and courtesy were I to omit expressing my
debt to many colleagues, past and present, including Dr.
Tryphena Humphrey, Dr. Ira D. Hogg, and the staff of the
Elizabeth Steel Magee Hospital. The work more specifi-
cally presented in the second and part of the third chapters
owes much of its financial support to grants from the Pen-
rose Fund of the American Philosophical Society, the
Carnegie Foundation of New York, the University of
Pittsburgh, the Sarah Mellon Scaife Foundation of Pitts-
burgh, and the late George Tallman Ladd, also of Pitts-
burgh.
It is a pleasure to express my indebtedness for many
courtesies to Dr. and Mrs. Paul G. Roofe, and to the staff
and graduate students in the Department of Anatomy at
Kansas. I also wish to thank Mr. Ivan Hird, whose ability
as a projectionist made the illustrations for the lectures
flow smoothly and accurately.
Contents
PREFACE v
I. FETAL ACTIVITY IN
INFRAHUMAN VERTEBRATES 3
I I . THE SEQUENCE IN HUMAN
FETAL ACTIVITY 54
I I I . THE SIGNIFICANCE OF STRUCTURAL
AND FUNCTIONAL INTERRELATIONSHIPS IN
PRENATAL ACTIVITY 87
NOTES 121
ACKNOWLEDGMENTS 122
REFERENCE LIST 123
INDEX OF NAMES 137
SUBJECT INDEX 139
Illustrations
1. Early flexure stage of Amblystoma 16
2. Coil stage of Amblystoma 16
3 . S-response of Amblystoma 16
4. Diagram of the nervous mechanism underlying
the early behavior pattern of Amblystoma . . . . 18
5 . Diagram of the caudal extent of the ophthalmic and
maxillomandibular fibers in the spinal tract of
the trigeminal nerve in young human embryos . . . 9 5
6. Diagram of the probable reflex pathway for contra-
lateral neck and upper trunk reflexes in the early
human embryo 9 8
Tke Prenatal Origin
of Bet a v i o r
Ckapter 1
Fetal Activity in Infraliximan
VerteLrates
B
EHAVIOR is A fundamental characteristic of all ani-
mals, whether adult or developing, unicellular or
multicellular. In essence, behavior is the sum total of
the adjustments made by the organism to changes in its
internal or external environment. The activities of the
mechanisms involved in restoring the dynamic balance of
the organism as a consequence of the environmental
changes give rise to bodily activity, which is the externally
visible or overt expression of the functional capacity of the
whole organism. In all but the simplest organisms, this ac-
tivity is produced primarily by the neuromuscular system,
but this overt behavior is also influenced by the activities of
other systems of the body, which constitute the organism's
internal, invisible, or covert behavior. The interacting in-
fluences of all the various organ systems of the individual
are infinitely complex, so that at any given moment the be-
havior of the whole organism tends to be different in Icind
and extent from the sum of the activities of its separate
parts.
Structure and function are directly and inseparably re-
lated in any living organism. This is as true of one which is
developing as it is of the adult. In the embryo, therefore,
when any organ reaches a level of development and differ-
entiation consonant with function, it begins to function,
provided the environment is appropriate. The qualities of
embryonic functioning of an organ may not be those ex-
[ 3 ]
Prenatal Origin of Behavior
[ 4 ]
hibited when its development is complete. Furthermore,
all organs do not reach a functional state in their develop-
ment at the same time. Therefore, the behavior of an or-
ganism attains the adult form characteristic of each species
only as it approaches maturity.
An embryo develops morphologically in an orderly, se-
quential manner characteristic of the species. Thus, given
an appropriate environment, an organism's behavior also
develops in an orderly, sequential manner which, again, is
characteristic of its species. However, the morphological
development of all vertebrates follows a fundamentally
similar sequence. Hence it would be surprising if there
were not a similar fundamental sequence of some type in
the development of behavior throughout the vertebrate
series. A priori, then, one might expect the developmental
sequences of behavior in the different vertebrates to pos-
sess certain fundamental similarities, but to differ in par-
ticulars in different species.
It is this thesis which is to be considered in some detail
in the first two chapters. The problem, then, is to survey
the origins of behavior, more particularly of the externally
visible overt behavior, in developing organisms throughout
the vertebrate phylogenetic series. The hope is to establish
the points of similarity and of difference between them, if
either exist.
The earlier recorded data on embryonic activity con-
sisted largely of casual and uncontrolled observations with-
out experimental attempts to analyze the nature of the
movements seen. Although one cannot merely write off
earlier observations as worthless, for the observers were
both keen and reasonably accurate, nevertheless they were
Fetal Activity in Infrahuman Vertebrates
[ 5 ]
pioneering efforts in a little-understood field and less valu-
able than those which came later.
It was not until the work of Wilhelm Preyer (1885)
that real organization and a scientific approach to the sub-
ject appeared. Preyer's interest apparently was stimulated
by observations of his own child (1882). However that
may be, he gathered together the work of his predecessors
in the field and added materially to the then existing state
of knowledge by his own wide and exact observations. His
work will be noted in the pages which follow.
Important as the work of Preyer was, and there can be
no dispute on that point, it was George Ellett Coghill who
first placed a firm foundation under the study of embryonic
activity. CoghilFs work was published in an extraordinary
series of 65 papers and one book, from 1898 to the time of
his death in 1941. Two of his unpublished manuscripts
were posthumously edited by C. Judson Herrick
1
and pub-
lished. This series of studies laid a solid foundation for the
correlation of the development of behavior with the asso-
ciated changes in the nervous system for the salamander,
Amblystoma. Coghill's work not only pointed the way for
similar studies on other forms, but also stimulated many of
his students and some of his associates to continue such in-
vestigations.
aje # aje >Je s{e #
One of the first muscular activities exhibited in the ver-
tebrate embryo is the heart beat. The early activity of the
heart in vertebrates has been demonstrated to be myogenic
in nature in many forms, that is to say, the rhythmic con-
tractility of its musculature is a property inherent in the
cardiac muscle cells themselves. Only later in embryonic
development do nerves, not yet developed when the car-
[ 6 ]
diac beat begins, control the activity of theheart. The beat-
ing of the heart is not ordinarily considered a component
of developing behavior. Certainly the heart beat is not a
part of the overt behavior of the organism, but an excellent
case can be made for its inclusion among the elements of
the internal, or covert, behavior of each animal form. It is
mentioned here, not to press the latter point, but because
similar myogenic movements occur in the trunk muscu-
lature of certain forms shortly to be considered in a neces-
sarily rather superficial review of the development of be-
havior in the infrahuman vertebrate phylogenetic series.
There are three types of movement exhibited by mus-
clesmyogenic, neurogenic, and reflexogenic. As already
noted, myogenic activity originates within the muscle tis-
sue itself, without benefit of a nervous or other impulse.
Actually the term "myogenic" is used to denote two quite
different types of muscle movement. One of these is ex-
hibited spontaneously and, usually, rhythmically. The
myogenic heart beat and the rhythmic activities of the
axial musculature of fishes are examples. Such myogenic
activity is an inherent property of the musculature. The
term is also sometimes used, unfortunately and probably
inaccurately, to denote muscle movements in response to
electrical or mechanical stimulation. In such cases there is
a stimulus to the muscle, so that its activity does not have
its genesis within the muscle cells as an inherent character-
istic of the tissue. When used in that sense in this contribu-
tion, the word will be enclosed in quotation marks.
The motor neuron, particularly in embryos, may dis-
charge impulses into muscle tissue, either spontaneously
or under the influence of stimuli from outside itself. Such
action of motor neurons in activating muscle is termed
Fetal Activity in Injrahuman Vertebrates
[ 7 ]
neurogenic. Again, the term is used for both the inherent
spontaneous type of motor discharge and that originating
from stimuli to the motor nerve.
In reflexogenic activity, a complete reflex arcreceptor
(sensory) neuron, usually one or more intercalated neu-
rons, and an effector (motor) neuronis involved. Motor
neurons appear to establish functional connection with
their muscle, ontogenetically, before their sensory or inter-
mediate connections are established. Thus neurogenic ac-
tion ordinarily ceases when the reflex arc is completed, but
it will become evident that in some forms an extension of
true spontaneous neurogenic activity may overlap the es-
tablishment of reflexogenic activity.
FISHES
Both of the principal investigators of activity in very
early selachian embryos, Stewart Paton (1907, 1 91 1 ) and
P. Wintrebert (1920), have given excellent descriptions of
the rhythmic contractility of the axial musculature ex-
hibited in these forms. Wintrebert's contribution, perhaps
because it succeeded a considerable period of study of the
earliest activity in amphibian embryos, offers somewhat
more detail and a more complete picture of activity se-
quences. He has described the side-to-side spontaneous ac-
tivity of the trunk of one of the cat sharks (Scyllioihinus
canicuh) as first appearing near the end of Stage G, as de-
scribed by Balfour (1874,1878). At this stage (G), only
one branchial pouch is present. The activity begins as a
property of the myotomic tissue, that is to say as a true myo-
genic phenomenon, of those myotomes capable of muscu-
lar contraction. The "aneural" (true myogenic) character
of these movements was proved when Wintrebert re-
[ 8 ]
moved the spinal cord without interrupting the activity.
The first movement to appear is often a unilateral flexion
of the trunk, moving the head toward either right or left,
and a return to the midline. Soon the movement becomes
bilateral, first to one side, then to the other, usually as two
bilateral swings of the head, each followed by a pause. Be-
cause the attachment of the embryo to the yolk sac offers
resistance to rotation, the trunk passively returns to its
original position during the pause.
The first contractions appear in the more cephalically
located myotomes, those first formed in development.
Gradually, myotomes behind these begin to participate in
the myogenic activity. Without entering into all the de-
tails furnished by Wintrebert, one may summarize the
characteristics of the side-to-side myotomic contractions as
follows: i ) each myotomic movement consists of a lateral
flexion of the head and its return to posture, followed by a
pause, then, in embryos in which a rhythm has been estab-
lished, a flexion to the opposite side, and its passive relax-
ation and pause; 2) the older the embryo, the greater is the
number of side-to-side flexions; 3) temperature plays an
important role, as the flexions succeed one another more
rapidly at the optimum temperature of 1 5
0
C. than at
higher or lower temperatures, and the time required for a
flexion, as well as the duration of the pause, also varies in
like manner; and 4) optimum conditions include, not only
a temperature of 1 5
0
C, but good aeration of the egg-case,
proper composition of the sea water in which it is retained,
and the absence of disturbing factors.
It is interesting to note that the heart beat in Scyllio-
rhinus does not begin until late in Balfour stage I, at which
time the myotomes begin to exhibit neurogenic activity,
[ 9 ]
again spreading caudally from those more cephalically lo-
cated. Still later, reflexogenic activity supervenes.
Paton's (1907, 1 91 1 ) factual results are, in the main,
the same as those of Wintrebert. However, he regarded his
results as evidence of the functioning of intercellular proto-
plasmic bridges capable of transmitting nerve impulses.
Hence, he viewed the movements as neurogenic in nature.
Hensen (1903) had earlier promulgated the idea that such
protoplasmic bridges existed and were, in fact, capable of
carrying impulses of a nervous type. Held (1909) had
modified the Hensen theory by what he believed to be a
demonstration of the formation of neurofibrils in these
bridges, which thus became true nerves, a position also
held by Graham Kerr (1912). The work of His (1887),
Cajal (1906), Harrison (1910), and others has quite defi-
nitely proved the nerve fiber to be a protoplasmic out-
growth of the perikaryon of the neuron and not a transfor-
mation in loco of protoplasmic bridges between body cells.
We may safely assume, then, that Paton's findings go to
strengthen the interpretation of these early spontaneous
movements as being myogenic in character.
Among very young teleosts, somewhat similar spontan-
eous movements have been observed by Tracy (1925,
1926) in the toadfish (Opsanus tau), the cunner (Tauto-
golabrus adspersus), and in several other forms; by Coghill
(1933b) in the killifish (FunduJus) and in the toadfish
(unpublished); and by Sawyer (1944) in Fundulus. Tracy
was at first of the opinion that these bilateral rhythmic
axial movements were neurogenic in nature, because he
believed that he could demonstrate in Opsanus early
motor paths in the medulla and cord, "but with the affer-
ent system largely undifferentiated during the earliest
[ 1 0 ]
stages of motor activity" (1926, pp. 352-53). He thought
that these motor nerves might possibly be stimulated to ac-
tivity by variations in the CO2 concentration in the em-
bryos (endogenous stimulation), as had White (1915) in
her studies on trout embryos. However, as a result of fur-
ther work, as yet unpublished, he became convinced that
they were myogenic in character.
Motility of the trunk first appears in toadfish embryos
with 19 to 22 somites in the most cephalic four or five pairs,
and the number involved gradually increases caudally. Un-
like the side-to-side rhythmic flexions of Scyllioihinus de-
scribed by Wintrebert (1920), the flexions of Opsanus are
not rhythmic, though they may be grouped, and produce a
somewhat coiled state of the body. When these coils are
performed alternately from side to side, Tracy termed the
responses "flutters" because of their irregular, nonrhyth-
mic nature.
This early myogenic activity persists through hatching,
to be followed by a similar cephalocaudally progressing
wave of reflexogenous activity. Whether a period of true
neurogenic movements exists between the myogenic and
reflexogenous periods is as yet uncertain. Tracy believes
this sequence is also true of cunner and Fundulus embryos.
Tracy's use of "endogenous" as synonymous with
"spontaneous" is unfortunate, but the present writerand
many othershave made the same error before the nature
of true myogenic or neurogenic movements was as well un-
derstood as it is today. Nevertheless, Tracy's 1926 paper
performed a major service in once more calling attention
to the work of many observers of the effects of CO2 in ap-
propriate quantities serving to facilitate, though not to
[ 1 1 ]
originate, activities of the nervous system, a matter to
which reference will be made in the third chapter.
Barron (1941) has pointed out that there is no evi-
dence that CO2 can stimulate a peripheral nerve in such a
manner as to initiate neurogenic action. However, if
Wintrebert (1920) is correct in his finding that the estab-
lishment of a motor nerve connection with a myotome
causes its spontaneous contractions to cease, such con-
tractions should be restored, if they are truly myogenic, by
the suppressing effect of CO2 on the nerve.
CoghilFs work on the earliest activities of Opsanus and
Fundulus gave support to Tracy's change in viewpoint (see
Herrick, 1949, pp. 96-97). Coghill (1933b; and see Her-
rick, 1949, pp. 264-69) divided the early activity of Fun-
dulus into five phases, beginning (phase A) with the first
localized myogenic contractions of myotomes which be-
came progressively active spontaneously in a general ceph-
alocaudal direction. The wave of myogenic activity, as it
spread caudal ward in phase B, was shortly replaced (phase
C) at the cephalic end of the body by another wave, this
one of responses to external stimulation, the initial sensi-
tivity becoming manifest in the area anterior to the eye.
In phase D, only the tip of the tail and the pectoral fin still
show localized spontaneous activity, which disappears in
the last phase (E). Almost identical results were secured
by Coghill in Opsanus. In both forms, the second or neuro-
genic wave replaced the myogenic in a caudal direction
until, as Coghill expressed it in a letter to Tracy (Herrick,
1949, p. 97), "The neurogenic system chases the myogenic
system off the end of the tail, the last of it appearing in the
movements of the caudal fin and at the same stage of de-
velopment in the pectoral fin also." Thus Coghill agreed
[ 121
with Tracy's suggestion that the earliest movements
throughout the somatic system were myogenic.
Reflexogenic activity follows the neurogenic in the be-
havioral sequence, but details are not available, since Cog-
hill's results were never published. Regarding this type of
activity, Coghill wrote me in July, 1933, as follows: "I
have found that the toadfish, in regard to the individuation
of limb reflexes out of a total pattern, fits absolutely with
Amblystoma, only the period is much longer between the
total pattern fin action and fin reflex than in Amblystoma.
9>
IgJusj^grimOTts on early Fundulus embryos, Cog-
hill used curare, which blocks the motor end-plates of
adult organisms. He found that the rhythmic spontaneous
movements continued unaffected, and accepted this as
clear-cut evidence of their myogenic nature. Barron
(1941) criticized, by implication, the curare experiments
of Coghill in discussing those of Angulo (1933). Barron's
point, with regard to the curare experiments with embryos,
was to the effect that the action of curare on immature
muscle had never been demonstrated. Its effect might be
the same on immature as on adult muscle, but, until proof
of this had been offered, any conclusion based on its use
that the spontaneous movements of embryos are myogenic
was subject to a Scottish verdict of "not proven." However,
recent work by Brinley (1951) indicates that curare, in the
strengths used by Angulo, Coghill, and Tracy, does act on
fish embryonic muscle as it does on that of adults.
On the positive side of the ledger stands the work of
Sawyer (1944), whose studies in correlating the type of
activity with the cholinesterase (ChE.) content of the
embryo, localized chiefly in future muscle and nerve, af-
ford strong evidence that the early spontaneous move-
[ 1 3 ]
ments in Fundulus are probably myogenic in nature.
Sawyer states (1943a) that "cholinesterase content is a bio-
chemical criterion of a phenomenon for which there is no
anatomical criterion, i.e., the attainment of functional
capacity in the neuromuscular apparatus" (p. 27). He has
pointed out that "spontaneous somatic movements begin
when there is practically no esterase present and eseriiie, a
ChE. inhibitor, does not affect the movements. They are
therefore interpreted as myogenic..." (1944, p. 82). This
opinion is reinforced by Sawyer's finding that eserine does
affect movement when the ChE. levels have risen materi-
ally, an occurrence which corresponds with the time true
reflexogenic activity is exhibited.
The development of behavior in fishes presents some
puzzling problems as yet incompletely resolved. To solve
them will require additional studies. Here, especially, there
is need for establishing distinctions between myogenic,
neurogenic, and reflexogenic activity. At the moment,
such distinction is by no means free from ambiguity.
Although the fishes occupy a lowly place in the phylo-
genetic series, it must not be assumed that they are all
necessarily primitive forms. Indeed, many are highly spec-
ialized, sometimes along peculiar lines which may well pro-
duce unusual developmental behavorial conditions, appar-
ent aberrant types in the theoretical fundamentally sim-
ilar sequence which we are examining. Also, there is in-
sufficient evidence on reflex behavior. Nevertheless, there
is much in the still incomplete knowledge of the develop-
ment of behavior in the fishes that sustains the idea of such
a fundamental sequence, probably at least as much as op-
poses the idea, even at this time.
AMPHIBIA
In the amphibians, at least in Amblystoma, we stand
on very sure ground. The work of Coghill (1909-1940) not
only laid the most solid foundation for behavioral develop-
ment existing today in any vertebrate group, but also
served to stimulate the majority of subsequent studies on
the development of behavior in all vertebrate forms. Con-
sequently, it is essential that we examine his contributions
in some detail.
Coghiirs studies on Amblystoma disclosed several
clearly recognizable stages in the development of its ac-
tivity following two others during which the embryo was
inactive. In the first of these two inactive stages, termed
the premotile stage, the embryonic myotomic tissue was
incapable of contraction. In the second, the nonmotile
stage, the myotomes were capable of contraction, but only
on direct mechanical or electrical stimulation. Such move-
ments were considered "myogenic," in that they did not
originate in the nervous tissue, but they were not spontan-
eous (seep. 6).
The first of the active stages of Amblystoma, termed by
Coghill the early flexure stage, represents the first of the
neuromuscular responses to external (exteroceptive) stim-
ulation. However, as shown by Preyer (1885) and by Cog-
hill, these movements also appear spontaneously. The first
area of embryonic ectoderm to become sensitive to stimu-
lation lies in the neck region over and anterior to the site
of later gill formation. As the ectoderm of all young em-
bryos is extremely thin, the method of stimulation is of
great importance, because too stiff a stimulator (esthesi-
ometer) may directly stimulate the underlying muscle tis-
sue either by pressure or by actual penetration. Coghill
[ 1 4 ]
[ 1 5 ]
used a human hair which he stroked lightly over the sensi-
tive area. It can thus be seen that the stiffness of the stimu-
lator must be correlated with the thickness of the epider-
mis of the embryo stimulated.
When the sensitive area of ectoderm in an Ambly-
stoma embryo in the early flexure stage is properly stimu-
lated with a human hair the response is characteristic and
stereotyped. "Stereotyped" is intended to convey the con-
cept that each response is, within the general limits of
variation of biological processes, practically identical with
every other, so that they are patterned and constant within
that pattern. The greatest variation found by Coghill con-
sisted in occasional flexions toward the side stimulated, al-
though the almost constant finding was a contralateral
flexion.
The early flexure response to stimulation within the
sensitive area consists of a flexion in the neck region, so
that the head almost forms a right angle with the long axis
of the body (fig. 1) and in the vast majority of cases does
this toward the side opposite that stimulated. When this
response first appears, the flexure at the neck is somewhat
less than a right angle and the contracting myotomes are
few in number. As development proceeds, more myo-
tomes, located caudally from those initially contracting,
become functional and the flexure becomes more marked.
It must be borne in mind that the development of ac-
tivity is always a continuous process and that any division
into phases or stages is entirely artificial. The whole process
is a smoothly progressive affair and not a series of isolated
posturings. Consequently, the transition from the early
flexure to the coil, CoghilFs second stage, is characterized
by the spread caudally of the flexion, as more and more
FIG. I. Early flexure stage of Amblystoma. Outline drawn by pro-
jection of the original Swenson-Coghill film. (By permission of
the Wistar Institute of Anatomy and Biology.)
FIG. 2. Coil stage of Amblystoma. Outline drawn by projection
from the original Swenson-Coghill film. (By permission of the
Wistar Institute of Anatomy and Biology.)
FIG. 3 . S-response of Amblystoma. Outline drawn by projection
from the original Swenson-Coghill film. (By permission of the
Wistar Institute of Anatomy and Biology.)
[ 1 6 ]
[17]
myotomes become functional, until the entire trunk is
coiled to the contralateral side, with the tail overlapping
the head (fig. 2). The coil results from a briefly maintained
contraction of the myotomes of the side opposite to that
stimulated.
For the third stage, the S-reaction (fig. 3), new neural
mechanisms, in the form of collaterals, are thrown into
action. Although the appearance of the S-response is rela-
tively sudden, the collaterals which make it possible have
been developing toward a functional stage for some time.
Herrick and Coghill (1915) and Coghill (1916, 1924a,
1929) worked out the neural mechanisms for the various
stages. CoghiU's generalized diagram (1924a, fig. 43) is
here reproduced in a slightly modified form (fig. 4). It will
be noted that the afferent mechanism begins with general
sensory components carried in the fifth, seventh, and tenth
cranial nerves. Although the last two nerves named con-
tribute little in the early stages, the trigeminal fibers enter
the descending tract of that nerve, as will those from the
facial and vagus later. The trigeminal general sensory fibers
have early grown caudally to synapse with tract neurons of
the spinal cord. This is in agreement with Coghiirs (1902)
findings in adult Amblystoma. Stimuli perceived in front
of the ear are transmitted as impulses over these general
cutaneous fibers. Behind the ear, sensations are picked up
by the cutaneous elements of the Rohon-Beard cells, and
their impulses enter the ascending sensory path as it de-
velops. All impulses are transmitted across to the contra-
lateral side of the cord by the floor plate cells, shown in the
middle of the diagram, thus causing contralateral re-
sponses.
Essentially, the S-reaction is an outgrowth of the coil
FIG. 4. Diagram of the nervous mechanism underlying the early be-
havior pattern of Amblystoma. (After Coghill, 1924a, fig. 43.)
The afferent paths are given on the left side of the figure, con-
nected by six floor-plate commissural cells (center) to the motor
pathway shown on the right side. Description is in the text, page
17. Abbreviations used: CNS, outline of the central nervous sys-
tem; E, ear; GC V, general cutaneous component of the fifth (tri-
geminal) nerve which enters the descending tract of that nerve; LL
VII and LL X, lateral line branches of the seventh (facial) and
tenth (vagus) nerves, which contain general cutaneous compon-
ents, later to course with those of the trigeminal in the latter's de-
scending tract; RB, Rohon-Beard cells, mediating both cutaneous
and muscle sense.
[ 1 9 ]
response, but with a difference. As the coil develops by the
rapid caudalward progression of the motor component of
the reflex, a transfer of the wave from the contralateral to
the ipsilateral side of the trunk occurs in the neck region.
Thus, as the now slowed contraction wave passes caudally
on one side of the body, a new contraction begins in the
cervical area on the other side and progresses caudally in
its turn. This throws the trunk into the form of the letter
S momentarily, but as each contraction wave passes off the
tail on one side, another starts cephalically on the same
side, thus alternating with that of the opposite side (fig.
3 ) . The trunk, then, undergoes a series of sinuous move-
ments in a head-to-tail direction, which, when they be-
come sufficiently rapid and strong, move the embryo for-
ward in the early swimming stage. At this time, no limbs
are present.
At this juncture, it seems advisable to return to the
mechanisms of the earlier stages, for they represent the
basis for Coghiirs concept of the "total pattern" response.
There has been so much misunderstanding of what he
meant by that term that some further consideration should
be given to it here.
Coghill found that the sensitive area of the ectoderm
was at first strictly localized and that, on the motor side,
all of the myotomes which were capable of being excited to
contraction by their motor nerves participated in the re-
sponse. Inasmuch as only those in the cervical region were
so prepared for action at first, the early flexure stage re-
sulted, but as more and more neuromuscular mechanism
became functional caudally, the early flexure widened into
the coil. In other words, there was a total response of all
functional neuromuscular apparatus. Coghill termed it a
[ 2 0 ]
"total pattern" response. The total pattern persists far be-
yond these early responses, however, and in Amblystoma
remains dominant throughout the life of the organism.
As has already been noted, these responses in the nor-
mal embryo are true reflexes. Weiss (1941a, 1941b,), how-
ever, has demonstrated that amphibian embryos with the
dorsal spinal roots cut along both sides of the entire spinal
cord may swim in perfect co-ordination. Herrick and Cog-
hill at first (1915) inclined to the opinion that proprio-
ceptive stimuli played an essential role in producing the S-
reaction. Later (1926b) Coghill recognized that the
neural mechanism governing this alternation of contrac-
tion waves on the two sides of the body might lie within
the motor system entirely, though he believed the proprio-
ceptive field facilitated the duration of the response. It
would appear that, under normal conditions and in re-
sponse to exteroceptive stimuli, the responses are reflexo-
genic, but that, under abnormal conditions, activities may
be actually neurogenic in origin. This appears to be the
chief evidence of actual neurogenic activity in the Am-
phibia. It has not been demonstrated as a separate type, as
is the case, apparently, in fishes,
2
unless the early spontan-
eous activity should prove to be neurogenic, which seems
unlikely.
Following the early swimming stage, at about the be-
ginning of which the Amblystoma larvae ordinarily hatch,
any true staging is difficult The larvae then begin to de-
velop along multiple lines, one of which is toward feeding
and another toward walking. As the limbs develop, they
are at first used solely as a part of the trunk muscle system,
having no power of independent movement. Indeed, even
after specific limb movements appear, locomotion on land
[ 2 1 ]
continues to involve the trunk, and in swimming the ex-
tremities are streamlined alongside the body. As the gills
develop, they also move with the trunk before indepen-
dent movements are possible. Each new structural com-
plex, as it begins to function, is at first co-ordinated with
the trunk as a part of the total pattern, only later coming
to possess the capacity for independent motion. This
secondary acquisition of the capacity for local movements
Coghill spoke of as the "individuation of partial patterns
from the total pattern."
In CoghilFs opinion, the initial movements exhibited
are total patterns which expand by shunting into the neu-
ral mechanism previously completely integrated additional
elements. Integration is thus a primary phenomenon oc-
curring during neural development.
Much more might be said regarding CoghilFs work,
but neither space nor time permits. It is essential, however,
that CoghilFs fundamental conclusions be presented in
some detail to establish the basic tenets of his concept of
the development of behavior. We have already seen that
he believed his concept to hold good for the fishes, and it
will become evident in subsequent pages that many per-
sons, a goodly number of whom were not his students,
have come to the conclusion that it also applies to other
forms. Even those who have considered his ideas as having
no validity for so-called higher forms, accept them for
Amblystoma. Certainly no other series of conclusions has
been based upon such detailed and adequate evidence.
It now becomes desirable to present briefly other im-
portant studies on the development of activity in the Am-
phibia. Preyer (1885), in his remarkable book already
mentioned, presents his observations on the spontaneous
[ 2 2 ]
movements of frog (Ram) embryos within the egg. Little
has been said about spontaneous movements in Ambly-
stoma so far, but they occur, as noted by Coghill repeatedly
(1930b and elsewhere). Unlike this phenomenon in the
fishes and in the amniotic contractions in birds, sponta-
neous movements in the Amphibia are not myogenic, but
are certainly either neurogenic or reflexogenic. The whole
question of spontaneous movements is as yet incompletely
resolved. Such movements may be caused by unrecognized
stimuli of various types or they may arise, quite indepen-
dently of stimuli, as an inherent property of nervous sub-
stance, as has been pointed out by Weiss (1941a, 1941b,
1941c) and noted earlier. Preyer's work, unfortunately but
not unexpectedly, throws little light on this aspect of the
problem. Nevertheless, his observations on sequence of ac-
tivities are quite accurate.
Youngstrom (1938) presented the results of a careful
study of the activities of embryos of Rana, Bufo, Pseuda-
cns
7
and Acris. He found surprising agreement in the na-
ture (though differences in time consumed in attaining
it) of the activity sequence in these anurans, which he de-
tails as follows: 1) spontaneous bending of the head, char-
acterized by slow contraction and relaxation, at a time
when any form of external stimulation is ineffective in
causing a response: 2) deep pressure stimulation, suddenly
applied, presumably in the branchial area, caused responses
in which the body was bent, often ipsilaterally, into an arc,
which he regards as the equivalent of CoghilFs coil stage;
3) again with quick deep pressure, the appearance of a
double wave of contraction forming a kind of S-reaction,
the amount of yolk present in the shorter trunk of anurans
probably accounting for the differences in these last two
[ 2 3 ]
responses from those of Amblystoma.; and then 4) irregu-
lar swimming movements; later, 5) to light touch stimula-
tion, the S and, soon after, swimming movements oc-
curred, the sensitive area being at first over the side of the
head, gradually increasing caudalward over the trunk and
finally reaching the tail. Youngstrom concludes that "the
Coghillian sequence of developing behavior has been
found to apply, with only slight variation, to the Anura"
(1938, p. 372).
Wang and Lu (1940, 1941), working on Rana and
Bufo, also agreed with the general outline of activity as de-
scribed by Coghill for Amblystoma, but more exactly with
Youngs trom's work on Anura. They describe six successive
stages, viz: 1 ) nonmotile; 2) flexure; 3) S-reaction; 4)
"translator^ body movements"; corresponding to Young-
strom's "irregular swimming"; 5) controlled swimming;
and 6) the appearance and maintenance of the upright po-
sition when at rest, as well as in motion, with the definite
presence of righting reflexes. In addition to their observa-
tions on the sequence of appearance of responses in normal
embryos, they transected the central nervous system at va-
rious levels, with instructive results. Removal of, or section
behind, the forebrain or diencephalon had no effect on the
sequence of appearance of activity throughout the embry-
onic period. Transection at the caudal border of the mesen-
cephalon prevented development of righting reflexes or
maintenance of upright posture. Cord section at any level
arrested the appearance of swimming. Furthermore,
spinal, decerebrate, or mesencephalic tadpoles exhibited
spontaneous movements, although adult anurans, so op-
erated upon, do not.
Cholinesterase (ChE.) determinations for Ambly-
[ 2 4 ]
stoma, Rana, and Bufo have been made by Youngstrom
(1938), and for Amblystoma alone by Sawyer (1943a,
1943b) and by Boell and Shen (1950). Youngs trom found
a significant increase in ChE. concentration in all three
forms during the period of increasing activity. Sawyer re-
ported that a small amount of ChE. was present in actual
or presumptive nerve and muscle tissues even in premotile
embryos, that the quantity of ChE. increases in nerve and
muscle during the swimming stages and reaches its high-
est peak at the feeding period, when activity is greatest,
gradually falling off in amount to the adult level. Boell and
Shen found that ChE. first appears in the spinal cord at
about Harrison stage 36,
3
when the embryo is capable of
responding to tactile stimulation. Gradually, detectable
amounts of ChE. appear, caudocephalically, in the hind-
brain (stage 38), midbrain (stage 39), and in the forebrain
(stage 42 or later). They believe that the increase in ChE.
is correlated with functional differentiation rather than
with mere increase in size and that the amount present
forms a gradient with the highest concentration in the
cord, decreasing steadily in a cephalic direction.
The importance of these observations, even though
they do not always agree in details, cannot be overesti-
mated. Cholinesterase depolarizes the nerve fiber after pas-
sage of a nervous impulse over it. Its presence in appreci-
able quantity, therefore, indicates that nerve activity is at
least possible.
In summary, it may be said that the Amphibia present
a clear-cut case for the total-pattern idea of the develop-
ment of behavior. This is based largely upon the classic
studies of Coghill on Amblystoma, but is supported as a
principle by all other investigators on amphibian forms. It
Fetal Activity in lnjrahuman Vertebrates
[ 25]
would appear that there is reasonable agreement, although
with expected variations, in the nature of behavioral de-
velopment in the tailed and tailless forms. Furthermore,
Coghill's basic work is accepted by almost all those who
have worked in the field.
REPTILES
Very little work has been done on the development of
behavior in any reptilian forms other than the turtles. Em-
inert and Hochstetter (1811) on lizards, Valenciennes
(1841) on python, and Preyer (1885) on ring adders ap-
pear to have presented the only (and meager) data on
other forms. This is an area where a great need for facts
exists.
Tuge (1 931 ), working with terrapin embryos, observed
spontaneous movements before exteroceptive stimulation
was effective. When the embryos are about 6 mm. in
length, the first sensitivity to external stimuli appears on
the snout. The response is a contralateral flexion of the
head and neck. Very soon thereafter the response spreads
caudally to involve the trunk, tail, and extremities, but the
limbs are not themselves sensitive at this time and do not
move independently of the trunk until the embryos are
between 7 and 7.5 mm, in length. At this stage the extrem-
ities become sensitive to stimulation, and both the fore
and hind limbs begin independent movement at the same
time.
Coghill (see Herrick, 1949, pp. 100 and 253) had ob-
served the studies by Tuge and confirmed each of his
points.
Smith and Daniel (1946) have published a prelimi-
nary report of their studies on embryos of the loggerhead
[ 2 6 ]
turtle, Caxetta caretta. Their results are of considerable
interest. Before the shell is more than a soft, dorsally lo-
cated swelling, they observed a "mass movement" (total
pattern) type of activity at 12 to 14 days, in the form of an
abortive kind of C-response, the head flexing to the con-
cave side, both on stimulation and spontaneously. At this
time the limbs did not participate, but moved, by 18 to 21
days, with the trunk as part of the total pattern. Specific
reactions became evident at 22 days, when the shell was
present in sufficient mass to prevent trunk movements.
These specific responses occurred first in the eyelids,
mouth, and head, as a whole, to be followed at 24 or 25
days by independent limb movements.
Professor Smith has been kind enough to supply addi-
tional details on the site of effective stimulation and the
nature of the C-response. When first manifested, the most
effective site for the reception of stimuli is the integument
over the head, with no specific localization, and the base of
the flippers comes next. The earliest C-responses have no
consistent direction, ipsilateral or contralateral, in relation
to the side stimulated, but after five to seven days, when
the carapace has developed further, they tend to become
ipsilateral. At that time, head extension has replaced lateral
flexion, so that the head-trunk movements have become a
"dual twisting of the head and body."
Between 26 and 32 days, activities appeared in the fol-
lowing sequence: 1) co-ordinated swimming by limb ac-
tion, 2) snapping of the jaws, 3) nystagmic head move-
ments when the embryo was rotated, 4) righting move-
ments, and 5) crawling. As these turtles do not hatch until
the 45th to 50th day, all of these activities occurred in
embryos removed from the egg.
Fetal Activity in Injrahuman "Vertebrates
[ 27]
The evidence available on reptiles, and this is actually
limited to observations on turtles, indicates that these
highly specialized forms present a Coghillian type of se-
quence modified by the rapid development of the carapace.
It is indeed unfortunate that no detailed studies have been
made recently on lizard embryos which may prove the least
specialized of these interesting animals.
BIRDS
Bird embryos have intrigued observers greatly over the
ages. Among the earliest recorded studies are those of Wi l -
liam Harvey (1651), who observed movements on the 6th
day of incubation of the chick. Beguelin (1757) was pos-
sibly the first to record the peculiar myogenic movements
of the amnion, a nerveless membrane which pulsates more
or less rhythmically at an early time. Von Baer (1828) and
Remak (1854) observed similar movements of the yolk
sac of the chick, and Vulpian (1857) demonstrated that
they may subsequently be found also in the allantois.
Most of the other early observations are of little moment
for our purposes, but it might be mentioned here that
Home (1822) recorded the first limb movements, and von
Baer (1828) very carefully reported the activities of the
6th to the 16th day, which were confirmed by Remak
(1854).
The chick egg, like other bird eggs, affords deceptively
easy access to the embryo. The difficulties lie not in ex-
posing the embryo but in maintaining it in a normally
living state while under observation for any extended
period. These difficulties were recognized by various inves-
tigators, and some made efforts to overcome or avoid them.
To this end, both Vulpian (1857) andPreyer (1885) ob-
[ 2 8 ]
served the developing embryo through the unopened shell
by transluminating the egg. It is amazing that they se-
cured so many results by this method, for Vulpian saw
both the amniotic and bodily movements of the chick em-
bryo, and Preyer was able to list the sequence of a con-
siderable number of activities, albeit his timing of the eggs
was slightly inaccurate according to later observations. His
series of recorded activities presents a good continuous
pattern of developmental behavior which in many respects
suggests a Coghillian type sequence. Clark and Clark
(1914), working on lymph heart pulsations, confirmed
most of Preyer's observations.
The three important recent contributors to the study
of the activities of bird embryos are Kuo (1932 to 1939),
Tuge (1934,1937), and Windle and his co-workers (1934
to 1938). Kuo and Windle et al. worked on the chick,
Tuge on the pigeon.
Kuo utilized a very ingenious technic to make his con-
tinuous observations possible. After carefully chipping
away the shell from the large end of the egg where the air
space is located, he vaselined the membrane to render it
moderately transparent. Strong transamination of the
shell made possible not only direct observation of the em-
bryo, but also the taking of motion pictures of unexpected
clarity. At the same time, the egg could be incubated in the
usual manner. He used several thousand eggs in his studies.
Kuo reported that the chick heart begins to beat at
about 36 hours of incubation. As the heart increased in
size, its beat caused "vibrations" of the head and body by
66 hours, head lifting at 68 hours, head bending at 70
hours, trunk movements at 84 hours, and head turning at
90 hours. The complications attending analysis of the na-
[ 2 9 ]
ture of the movements observed as a result of the general
upheaval of the embryo by the heart beat are further in-
creased, from 86 hours on, by the appearance of amniotic
contractions. The difficulties of such an analysis are much
like attempting to determine whether movements ex-
hibited by a flexible object floating in the ocean during a
storm at sea originate within the object itself or are caused
by the environment. Kuo believed that the activity of the
heart and later of the amnion passively excites these spon-
taneous activities of the embryo.
In the latter part of the fourth day of incubation, Kuo
(1932a) states that reflexogenic activity begins as a spon-
taneous phenomenon, the movements of the amnion and
yolk sac serving as the source of stimulation. These reflexes
increase in frequency at first with increase in age. The se-
quence of appearance of reflex activities is cephalocaudal
in direction and their character tends to parallel the earlier
non-reflex movements until after the 14th day, when the
myogenic activity of amnion and yolk sac, and the sponta-
neous reflexes excited by it, cease. The reflex activities ap-
pear in the following order: 1) head bending, 2) trunk
"bending," extension and twisting, 3) head turning, 4)
forelimb movements, then responses of 5) hind limbs, 6)
tail, 7) beak, 8) toes, 9) eyelids, 10) eyeballs, 1 1 ) swallow-
ing, 12) bill clapping, 13) trunk wriggling, and 14) trunk
rotation.
Although not too sympathetic to Coghiirs views, Kuo
himself (1932a) states that the earliest response pattern
found by him "agrees with Coghiirs ('29) observation."
However, Kuo fails to make careful distinctions between
the presumably myogenic early movements and later pos-
sibly neurogenic or definitely reflexogenic activities. The
[ 3 0 ]
reflexes, as their development is described by Kuo, do form
a Coghillian sequence, but the embryo's movements show
a progressive reduction in their frequency and amplitude
after the ninth day, at which time response to tactile stim-
ulation first appears. Responses to pressure were, however,
observed earlier (5th day) in some embryos.
Kuo (1939) was able to demonstrate the presence of
acetylcholine in chick embryos at 60 hours, but no evi-
dence of the presence of cholinesterase has been provided.
Kuo states that he could determine no evidence that the
presence of acetylcholine had any effect upon the develop-
ment of reflexes.
Preyer believed that the chick embryo spontaneously
exhibited a kind of C-shaped trunk flexion at the time
amniotic movements began, but Kuo fails to note any such
action. The latter recorded a series of spontaneous move-
ments as appearing in the following order: active limb
movements at 90 hours, when the head turned, also spon-
taneously; tail movements at 92 hours; neck flexion at 96
hours; independent, active, but irregular movements of
head, trunk, tail, and limbs at 1 1 5 hours; opening and clos-
ing of the bill at 7 days; and swallowing a day later.
The embryos exhibited responses to electrical stimuli
by 90 to 100 hours but did not exhibit responses to deep
pressure until about 144 hours. Local limb reflexes were
elicited at 155 hours, and respiratory movements began at
15 to 18 days.
Tuge (1934, 1937) worked out the sequence of ac-
tivity in the carrier pigeon. At 85 hours, the embryos begin
to exhibit generalized wave-like spontaneous movements,
believed to be myogenic in type because curare does not
abolish them. These persist until the sixth day. At 95 hours,
[ 31]
there are superimposed upon these spontaneous move-
ments others which are eliminated by curare. These move-
ments, spontaneous flexions of the neck and head, may be
unilateral or, in a few cases, bilateral and are considered by
Tuge to be neurogenic in nature. These spontaneous flex-
ions expand caudally to include the trunk (101 hours), the
rump ( 1 1 5 hours), the tail (120 hours), and the extrem-
ities with the trunk (125 hours) and, finally, independent
tail and limb movements (133 hours). This type of ac-
tivity ceases between 135 and 144 hours.
In the meantime, responses to stimulation of "snout"
or upper neck with a human hair begin at 123 hours in the
form of flexions of the head, neck, and trunk. The reflexo-
genous zone of the skin spreads in the following sequence:
"snout," neck, trunk, back, and hip by 133 hours. Simi-
larly, the reaction pattern expands caudally to hip with
trunk (125 hours), tail with trunk (130 hours), and ex-
tremities (depending on which is stimulated) with the
trunk (133 hours). Independent tail reflexes were observed
at 145 hours and of the extremities at 150 hours. Thus,
Tuge's results agree with the concept of a true Coghillian
sequence.
Another, but less extensive, series of observations on
bird embryos are those of Windle with his collaborators
(1934 to 1938), carried out on the chick. These investiga-
tors have presented analyses of the spontaneous motility,
and that in response to mechanical (tapping the embryo
or "flipping" the limb with a blunt fiber needle) and elec-
trical (faradic) stimuli.
The sequence of spontaneous activity is listed as given
below by Orr and Windle (19 34). The first time given rep-
resents the initial appearance of the particular type of ac-
[ 3 2 ]
tivity, the second that at which a majority of embryos ex-
hibit it. The sequence is: 1 ) ventroflexion of the trunk, 4%
to 5 Vi days; 2) lateral flexion of the trunk, 5
1
/
2
to 6 days; 3)
"swimming" type of activity, 5V2 to 6
1
/4 days; 4) move-
ment of the tail with the trunk, 5V2 to 8 days; 5) move-
ment of the extremities with the trunk, 7 to 8 days; 6) in-
dependent movement of the extremities, j
l
A to 9 days;
and 7) independent movements of the tail, 8 days in some.
It will be noted that the time of occurrence of these activi-
ties is later than that given by Kuo and by Tuge, though it
is sometimes difficult to match performance in the several
studies. This is, perhaps, only a semantic difficulty.
In response to mechanical stimulation of the types
used, generalized responses to trunk stimulation occurred
at 6 to 9 days; local wing reflexes to wing stimulation at 6
to j
l
A days; generalized "writhing" of the trunk to wing
stimulation at 6 J4 to 9 days; local leg reflexes to leg stimu-
lation at 7 to 8 days; generalized "writhing" of the trunk
to leg stimulation at 7 Vi days in some; and either local or
generalized response to "snout" stimulation at 7 to 9 days.
Here again, quite aside from the adequacy or inadequacy
of the stimulation used for analysis of the different types of
activity elicitable, the terms used to describe the responses
have a confusing, rather than a clarifying, effect. In re-
sponse to faradic current stimulation, local muscular con-
tractions were secured at 6 to 6
l
A days. That they were not
found earlier is puzzling, since ventral and lateral flexions
of the trunk occurred by 5 days.
Windle and Orr (1934) believe that when these trunk
flexions occurred spontaneously at about 5 days, the sen-
sory elements of the possible reflex arc were only "very
feebly developed," and that there were no association neu-
[ 3 3 ]
ronal connections present. Some 12 hours later, when
spontaneous bilateral trunk flexions were observed, the
ventral longitudinal pathway of the cord was connected
with the motor paths by collaterals into the mantle layer.
They report that at 6 days, when their type of "mechanical
stimulation" first became effective in the form of unilateral
wing reflexes, collaterals from the dorsal funiculus entered
the alar plate mantle layer, and they believe that this is the
channel of reflex action at this time. Although the spinal
accessory nerve (XI) is believed to function in the earlier
reflexes, afferent fibers in the dorsal roots are presumed to
take over behind the second cervical level.
The birds constitute a very difficult group in which to
study the sequence of activity in development. This is true
for several reasons, most notable of which is the constant
motion imparted to the entire embryo by the beating of
the relatively large heart and by the contractions of the
amnion and allantois. Presumably, except for the defi-
nitely myogenic character of the amniotic contractions, all
of the early spontaneous body activities are at least neuro-
genic, but proof is still lacking. Whether or not the se-
quence of activity is Coghillian in nature is strongly con-
tested by the several investigators. What is needed here is
further study unbiased by preconceived ideas as to what
the results should demonstrate. Such studies can be made.
The difficulties to be overcome are not as formidable as
those encountered in the work on mammalian embryos.
INFRAHUMAN MAMMALS
Historically, except for the work of Preyer (1885) and
Lane (1 91 7), systematic studies on the developmental se-
quence of activity in the infrahuman mammals began
"Prenatal Origin of Behavior
[ 3 4 ]
while Coghiirs investigations on Amblystoma. were still in
progress. The work of Swenson and Angulo. actually the
first of the more recent mammalian behavioral studies,
were carried out at Kansas under Coghiirs direction. Only
Tracy's studies on the fishes were contemporaneous.
Preyer (1885) published the results of a rather exten-
sive series of observations on the embryonic development
of activity in the guinea pig. Although his observations
were conducted with care and a keen eye, he overlooked
the rapid cooling of the embryo or fetus. This is a most
important consideration which may be overcome quite
readily and which causes a rapid rise in the threshold to
stimulation and earlier failure of the neuromuscular re-
sponse of the organism being observed. The other factor
was the hypoxic condition permitted to supervene. In
some forms, hypoxia may be overcome only with great dif-
ficulty and, per se, is far less important than cooling. How-
ever, failure to take these factors into consideration tends
to detract from the significance of the results secured.
The work of several other investigators should be men-
tioned here, although, for one reason or another, their
work is of less importance for the purposes of this review.
Lane (1917) studied the development of function, in rela-
tion to structure, of the special senses of the albino rat.
Avery's (1928) studies on guinea-pig fetuses were limited
to older ages, as was the work of Tilney and Kubie ( 1931)
and of Tilney (1933) on the interrelation of brain struc-
ture and level of activity exhibited in cat and rat embryos.
Originally projected to include a study of embryos of the
opossum, the guinea pig, and man as well, the investiga-
tions were incomplete because of the death of the senior
investigator.
[ 3 5 ]
The initial modern work on the Mammalia was insti-
tuted by Coghill in the Department of Anatomy at Kansas
as a part of his program of covering most of the vertebrate
classes, and at about the time Tracy began his studies on
Opsanus. The plan was for two of his graduate students to
attack the problem in rat embryos, one, Swenson, to study
the physiological responses, the other, Angulo, to do the
morphological part of the investigation.
In his University of Kansas Ph.D. thesis (1926),
4
un-
fortunately never published, Swenson detailed a series of
six successive response types exhibited by the rat embryo
between the initiation of reflex activity on the 16th day
and the 18th day after insemination. This series began with
a contralateral flexion of the neck, trunk, and rump at 378
hours, passed through a stage of lateral flexions of this type
alternating from side to side, followed in turn by forelimb
flexion and extension, hind limb abduction and adduction,
dorso-ventral extension-flexion movements, first apparent
in the head and then extending to the trunk, and rump ro-
tation with tail and thigh extension, the last on the 18th
day. In a report to the American Association of Anatomists
in 1928, Swenson somewhat modified his succession of
trunk movements into a series as follows: 1 ) unilateral
trunk flexions, 2) head extension, 3) ventral rump flexion,
4) head rotation, and 5) rump rotation.
It must be borne in mind that these observations were
of a pioneering nature. Until experience has been gained
by a long series of observations, it is difficult to analyze
what has been seen. Furthermore Swenson, in addition to
the usual method of stroking the embryo with a hair,
adopted two types of stimulation which complicate in-
terpretation of the reactions. One of these was to clamp the
[ 3 6 ]
umbilical cord. This procedure causes frantic bodily move-
ments in young subprimate fetuses, though in human em-
bryos and fetuses no such response follows clamping of the
umbilical cord. The other was the use of stiff stimulators
to "flip" a limb. By forcibly extending a limb and then al-
lowing it to return to position with a sudden jerk, very
early limb movements secondary to the "flip" may be se-
cured. Although such movements can be elicited at about
the time activity in response to true exteroceptive stimuli
occurs, Swenson believed them proprioceptive in nature.
Coghill, however, was of the opinion that they were the re-
sult of direct muscle stimulation. Actually, either explana-
tion may be valid, in the absence of morphological evi-
dence. To set in motion stretch reflexes, usually considered
to be bineuronal, requires far less muscle elongation than
that provided by the limb-flip.
In 1929, Swenson, who had left the Wistar Institute
for the Department of Anatomy at Pennsylvania, adopted
a "simple movement" plus "simple movement" idea with
the statement, "Each simple movement makes its first ap-
pearance in a definite order with relation to the other
simple movements." He then believed that the order of
appearance of movements had to do with three successive
categories, namely: progression, respiration, and ingestion
of food. Unfortunately, this seems a marked oversimplifi-
cation not borne out by the work of others.
After Swenson left the Wistar Institute, Angulo took
over the survey of physiological stages (1932-1951). His
studies on the morphological aspects of the nervous system
were published from 1927 to 1932. In his studies on the
development of motility in the rat, Angulo (1932b) estab-
lished four major phases: 1) the nonmotile, 2) the "myo-
[ 3 7 ]
genie/' 3) the neurogenic, and 4) the reflex. The non-
motile phase ends at about 360 hours (15th day), to be
followed by the "myogenic." The "myogenic" responses
are not spontaneous, but are manifested on mechanical or
electrical stimulation of the muscle tissue. The "myo-
genic" phase is very brief, lasting in general not more than
10 hours and followed by another, even briefer period of
neurogenic activity.
Between the 378th and 380th hour, reflex activity ap-
pears in response to stroking the snout region with a light
hair. The first response is a contralateral flexion of the
trunk, especially in the neck and upper trunk region. In
all, Angulo was able to establish a series of 30 responses
(tables 1 and 2). Examination of these responses demon-
strates that they form a well-integrated Coghillian se-
quence, even though in some details it differs from that
exhibited by other mammalian forms, as we shall see. This
is, however, to be expected and it is impossible to overem-
phasize the fact that each species of vertebrate has its own
variations in sequence, insofar as details are concerned.
Pankratz (1 931 ), in the course of a study on the de-
velopment of the adrenal gland in the rabbit, reported in a
preliminary contribution six steps in the development of
its overt behavior. These were: 1 ) lateral flexion of the
neck and upper trunk (15 to 16 days), 2) ventroflexion of
the neck and upper trunk with some movement of the
forelimbs as part of the trunk response (17 days), and by
the 20th day, 3) opening and closing of the mouth, 4) in-
dependent movements of the forelimbs, 5) hind limb flex-
ion, and 6) lateral flexion of the entire trunk. So far as in-
formation is given, we have here a reasonable Coghillian
TABLE I
Index to nature of each response secured by A. W. Angulo, 1932b,
listed in table 2.
(By permission of the author and of the Wistar Institute of
Anatomy and Biology.)
0, non-motile
1 , lateral flexion of the trunk
2, lateral flexion of the trunk with
movement of the fore limbs
3, lateral flexion of the rump
4, extension of the head
5, extension of the head with open-
ing of the mouth
6, extension of the head with open-
ing of the mouth and protrusion of
the tongue
7, lateral flexion of the rump with
movement of the hind limbs
8, ventroflexion of the trunk and
rump
9, independent movement of the fore
limbs
1 0, maintained contractions
1 1 , contraction of the abdominal
muscles
1 2, extension of the rump
1 3, flexion at the elbow and wrist
associated with trunk movements
1 4, attempt to assume "the optimum
physiological posture"
15, rotation of the trunk and rump
1 6, independent movement of the
hind limbs
1 7, extension of the head and rump
with kicking of the hind limbs
18, independent opening of the mouth
1 9, independent extension of the
hands
20, independent flexion of the hands
21 , specific reflexes
22, movement of the tail
23, flexion at the ankle
24, wrinkling of the skin
2 5, flexion of the hip
26, movement of the toes
27, independent movement of the feet
28, independent movement of the
tongue
29, wrinkling of the snout
30, independent active closing of the
mouth
sequence, though one differing in details from that de-
scribed by Angulo in the rat.
Coronios (1933), working on cat embryos, observed
still other variations in the order of development of ac-
tivity (table 3). His method of stimulation, a lightbrush,
is good but allows less information concerning exact local-
ization than does a light hair. The sequence of appearance
of reflex activity presented by Coronios actually lends con-
siderable support to the Coghillian total pattern idea, al-
though neither Coronios nor Carmichael, under whose di-
rection the work was done, was sympathetic to the total
pattern concept. One of the difficulties in the interpreta-
tion of these results is again semantic. "Retraction" of a
limb can be its extension, abduction, or adduction. Ap-
[ 3 8 ]
Percentage of responses at indicated ages in the rat embryo and
fetus, from Angulo, 1932b. (By permission of the Wistar Insti-
tute of Anatomy and Biology.)
K R A C T t OU l
14 D A Y 15 DAYS 10 DAY! 17 DAY* 1ft D A Y *
I M > A Y
20 DAYS 21 DAY S
0 100 68 29
1
32 71 71 36 7
2 59 92 66 52 52 100
3 43 54 35 13
4
7 81 96 89 72
100
35 74 58 27
0
10
2
43 20
7
5 49 53 67
8 8 54 38 31 28 45
9 If 42 42 26 20 80
10 2? 8 77 89 72 45
11
54 38 31 28 45
12 30 47 90 75
100
13
IK
14 9 23 24 65 45
15 21 34 18
16
10
16
20
27
100
17 5 53 H9 70
100
1
30 20
*.
45
1 52 79 96
100
20
48 78 96 100
21
31/ 79 9<
100
22
11 10 40
23
It
24
4 *>
25
3
2
27
4 65 65
28
3f 52 65
29
51 83 00
30
13 42 100
[ 3 9 ]
parently only training in human anatomy, with its exact
terminology of the nature of different movements, affords
the desirable background for avoiding this difficulty.
Carmichael (1934) made an extensive study of the
guinea pig embryo from 27 to 67 postcopulation days for
areas sensitive to various types of stimulation. He used
punctiform stimulation and stroke with a "pliable" hair,
pressure with a "strong" bristle or blunt probe, pricking
TABLE 2
TABLE 3
Table of responses in the cat fetus, from Coronios, 1933. (By per-
mission of The Journal Press.)
with a sharp needle, "pinching skin and muscle" with for-
ceps, exploration of the mouth with a probe, passive move-
ment of the extremities, electric current, noise, light, water
drops, etc. Such a battery of methods of stimulation has
rarely been used on any organism. Local muscle contrac-
tions were secured at 27 days of copulation age to faradic
current stimulation, before any spontaneous movements
(28 days) were executed and before responses to extero-
ceptive stimulation could be secured at 31 days.
The sequence of spontaneous movements executed by
[ 4 0 ]
[ 4 1 ]
the embryos in amnio between 28 and 31 postcopulation
days has close similarity to that for reactive behavior as de-
scribed by Pankratz for the rabbit and Angulo for the rat.
CarmichaeFs sequence for exteroceptive responses is di-
vided into five stages, in the first of which "certain aspects
of what will later form the typical 'pattern' of the gross
responses from that area are released. By 'gross responses'
is meant large limb-muscle and trunk movements" (1934,
p. 430). However, he found that limbs moved at an early
time toward the area stimulated, in conjunction with other
bodily movements. This is an almost unique finding and is
difficult to understand, as early movements have seldom
been observed to have an apparently adaptive application
of this sort. Most investigators have described the early
trunk movements as being of a type that is of the nature of
an avoiding response, although there are exceptions, as will
be noted shortly.
In Carmichael's second stage the "pattern of gross
movement which appeared weakly and incompletely in
Stage A is now stronger and in many cases more 'adap-
tive/ " and specific reflexes begin to appear in his third
stage. In the fourth stage, the "larger pattern" has become
less complete with the increase in the "local responses" al-
though it "can still be evoked." In his fifth stage, previously
effective receptor areas tend to produce only "fast and pre-
cisely localized" responses or previously distinct patterns of
reactivity tend to merge into newer and larger patterns of
the entire organism. Carmichael has been contra-Coghil-
lian in viewpoint because he believed the earlier extero-
ceptive responses to be far more complex in character than
described by Coghill.
Carmichael was kind enough to demonstrate to me
[ 4 2 ]
one of the relatively early responses of the guinea-pig em-
bryos. A stiff bristle was inserted into the external auditory
canal, the region regarded by Carmichael as the first sensi-
tive skin area in the guinea pig. A marked adduction move-
ment of the ipsilateral forelimb resulted, accompanied by
a slight neck flexion. It is easy to attribute responses to
what some investigators, myself included, would regard as
abnormally stiff or sharp stimulators, but the human em-
bryo, the only mammalian form with which I have had
extensive personal experience, does not respond to any
stimulation in this area at a comparable stage of develop-
ment. Except for these peculiar and unique movements,
which may be a species difference, the sequence as worked
out by Carmichael for the guinea pig has greater resem-
blance to the total pattern than to the additive idea orig-
inally put forward by Swenson. It is also true that the ap-
parent greater complexity of early movements may be a
matter of interpretation based upon the greater early par-
ticipation of the forelimbs in mammals, in particular, than
in amphibians. It is certain that adduction of the forelimbs
occurs earlier in the guinea pig than it does in the human
embryo and this may well be the basis of their "adaptive"
movement toward the site of stimulation.
Bridgman and Carmichael (1935) found that re-
sponses to electrical stimulation preceded other types of
activity which they regarded as reflexes. Slightly later (at
622 hours), the early guinea-pig fetuses responded to stim-
ulation of the intact amniotic sac, either by exertion of
pressure on it or by lightly striking its surface, by indepen-
dent head or forelimb movements. Some 10 to 14 hours
later, spontaneous movements of the same type appeared.
Lateral trunk flexion often with neck flexion to form a C-
[ 4 3 ]
type response, though at other times independently, could
be elicited at 6 4 3 hours. These investigators believe that
independent movements of a reflex nature occur at all
times even in 712-hour fetuses, the oldest re-examined by
them in this study.
Discussion now turns to the work of Windle and his
associates (1930-1950). It is this group which ultimately
espoused Swenson's idea of the origin of mammalian em-
bryonic reactivity by the successive addition of "simple"
reflexes. However, Windle and Griffin (1 931 ), in their
studies on spontaneous and reactive movements in the cat
embryo, reported a sequence very close to Coghillian in
its characteristics (table4).
An element of doubt concerning the correctness of
these earlier results began to appear in the 1933 paper by
Windle, O'Donnell, and Glasshagle, which was also on the
cat embryo. The earliest spontaneous movements observed
at 24 days were first in the forelimb, later in the neck.
However, they still considered the "mass movements" of
the head, neck, and limbs to be totally integrated from the
beginning. Tapping of the amnion and stimulation with a
fine steel needle or a cactus thorn were the methods used.
Both needle and thorn might be considered ill adapted to
avoidance of direct mechanical stimulation. Some of their
statements are of interest in this connection: "Simple, lo-
cal, unit muscular contractions can be elicited a day earlier
than spontaneous mass movements/' but these may not
be reflexes; localized stimulation elicited "isolated move-
ments locally and reflex contractions of muscles in regions
rostrad. However, reflexes farther caudad never ensue until
after the expanding massive total pattern (locomotor re-
TABLE 4
Table of activities of the cat fetus, from Windle and Griffin, 1931.,
(By permission of the Wistar Institute of Anatomy and Biology.)
[44]
[ 4 5 ]
action) has extended below the points stimulated" (p.
540).
Windle, Orr, and Minear (1934), in their further
studies on cat embryos, found that direct muscle stimu-
lation could cause contractions at 23 days of copulation
age and that the first reflexes appeared in some cases on the
24th day. Furthermore, limb movements followed 'pres-
sure" stimuli earlier than they could be elicited by faradic
shocks or by touch, but the reverse was true of head exten-
sion.
In his 1934 P
a
P
e r
> again on cat embryos, Windle
(1934b) made known his complete reversal to the "simple"
reflex idea in the following statements: "Morphologically
and physiologically, there is no evidence that the concept
of a gradually expanding total pattern completely inte-
grated from the beginning (Coghill, '29) can be applied
to the first leg reflexes of the cat embryo. A series of uniseg-
mental reflex arcs has been demonstrated at the time re-
flexes appear. The location of these in the upper brachial
segments is correlated with the physiological observations.
. . . Of course the local reflex arcs must eventually be inte-
grated with each other as well as with more distant centers
in the central nervous system" (1934b, pp. 501-502).
These items will be discussed later in some detail.
Windle now turned his attention to an attempt to
assay the effects of removal of cat embryos from the uterus
in an effort to determine the "behavioral capabilities" of
embryos after delivery, which are manifestly not those ob-
taining in the normal intra-uterine environment. This was
begun in a communication by Windle, Minear, Austin,
and Orr (1935). Nonmotilestage embryos were subjected
to faradic stimulation. Under such stimulation, as these
[ 4 6 ]
authors pointed out, the responses were always maximal
and without latent period. The muscle tissue was held in a
tetanic stage during the shock and relaxed slowly when it
ceased. There was no refractory period, contractions being
obtainable at will. The response was always localized and
depended on the position of the electrodes. Furthermore,
the physiological state of the embryo was without effect,
the responses secured from dying or recently dead embryos
being of the same character and elicited with the same
facility as from those freshly exposed.
Attention was then turned to reflex behavior, and the
time of eliciting the first reflexes was advanced from the
previously stated time of 24 days to the 16th day. The
methods of stimulation were "flipping" of the forelimb or
tapping the embryo or amnion. Limb "flipping" or snout
tapping caused movements of the limb or head, respec-
tively, the former usually being independent of trunk
movements. Each of the qualities of responses secured by
the means mentioned were the exact opposite of those ob-
tained by faradic current, viz., they were quick, nonmaxi-
mal, with definite latent and refractory periods, nonte-
tanic, stereotyped without relation to the site of stimula-
tion, and elicitable only from freshly exposed embryos. An
interesting note is that "all reflexes seemed to require
strong stimuli administered over relatively large areas in
their incipiency" (p. 174). The question of strength of
stimuli required for early reflexes is another aspect to be
considered later.
In 1936, after having worked in Sir Joseph Barcroff s
laboratory at Cambridge University, where sheep embryos
were being studied, Windle published a short article on
the activities of this and other forms. He stated that the re-
[ 47]
sponses secured in response to forelimb "flipping" and tap-
ping on the intact amnion were probably "local, isolated
movements, independent of and not as yet incorporated
into any behavior pattern; that each involves the nervous
as well as the muscular system and probably more than one
neurone. I suggest that these embryonic movements are
essentially reflexes of an elementary sort. They should not
be confused with myogenic contractions, from which they
differ in a number of respects" (p. 32P). "I invite a con-
sideration of the probability that mammalian somatic be-
havior has its genesis in simple, non-integrated, spinal type
reflexes" (p. 33P).
From 1938 to 1942, Windle and his collaborators
turned their attention to the subject of anoxemia and its
effect upon embryonic activity. They developed, early in
this study, the thesis that all activities of early embryos
examined more than a minute or so after the opening of
the uterus, whether or not the embryo was still attached
to a normal placental circulation, were abnormal by reason
of the hypoxic condition produced. In 1944, Win die's
"synthesizing article" presented these concepts in some
detail. Here he re-emphasized the rapidity with which hy-
poxia abolishes early limb reflexes and transforms the orig-
inal homolateral responses to a contralateral type. He thus
believed that all Coghillian type responses (contralateral
total pattern responses) in mammals were due to the hy-
poxic or asphyxial conditions induced by opening the
uterus.
The work of Windle and his associates has been pre-
sented in such detailed review because it represents the
chief challenge to the work of so many others and because
a number of his claims will be examined in the third chap-
[ 4 8 ]
ter. He began with a more or less Coghillian view but grad-
ually adopted Swenson's (1929) "simple" reflex additive
concept and his limb-flip technique. He ended with the be-
lief that the total pattern type of response was solely due
to the anoxic state of the removed fetus. These beliefs are
reiterated in his 1950 (1950b) paper.
From 1936 to 1950, Barcroft, chiefly with Barron, but
also with others, and Barron studied intensively the de-
velopment of motility in sheep embryos. Work on this
form had been published prior to 1936, but, because of the
use of anesthetics which affected the activities of the fetus,
it will not be considered here.
In 1936, Windle worked in Barcroft's laboratory, as
previously noted, and the paper published in that year by
Barcroft, Barron, and Windle largely reflects the views of
the third author. However, as Barcroft and Barron con-
tinued their studies, their results caused them gradually
to shift in the direction of a Coghillian sequence, and in
1939 they stated: "So far therefore as local and mass
responses are concerned in the sheep, local mechanisms do
appear to become segrated out from the total response in
the sense implied by Coghill in his account of the develop-
ment of behavior in Amblystoma" (1939b,p. 500).
Barcroft and Barron (1939a, 1939b) have summarized
their previous studies (1936, 1937) and noted the se-
quence of events in the sheep embryo (gestation period,
147 days). The first activities exhibited were "spontan-
eous," appearing on the 35th day in favorable cases and
consisting of lateral flexions of the trunk and head, pro-
duced by contraction of the long back muscles, and ac-
companied by abduction of the "poorly developed" ipsi-
[ 49]
lateral forelimb. The movements were quick in execution
and in return to posture.
At 38 days, the trunk was still flexed laterally, but the
head was now extended and both fore and hind limbs were
abducted. These "spontaneously" executed movements
occurred repeatedly, establishing a kind of rapid rhythm,
from side to side. At the same time, movements of the
liver appeared to indicate that the diaphragm was contract-
ing as a part of the "generalized" activity, inasmuch as
(1936) such diaphragmatic contractions could not be
elicited at this time by asphyxial conditions.
Between 41 and 47 days, limb activity began on an in-
creased tempo and amplitude, but associated with the
movements of the trunk. Eye movements, as part of the
"generalized" "spontaneous" activity, were observed on
the 50th day. When the head was moved suddenly in a
"clockwise" direction, the left eye was "just preceptibly"
moved downward.
These "spontaneous" movements, increasing in extent
and amplitude with age, were so called because no specific
stimulus could be assigned as their cause. However, be-
cause these "spontaneous" movements tended to disap-
pear if the uterus or amniotic sac were left undisturbed for
a period and immediately to begin again if either were
moved, Barcroft and Barron concluded that the move-
ment of uterus or sac caused afferent neuron stimulation,
at least in large part, if not entirely. They were unimpressed
by the notion that such movements were caused by "ac-
cumulated metabolites, principally CO2, " as they were
never able to produce these movements by inducing as-
phyxial conditions at this age.
In studying induced activity caused by faradic stimuli,
[ 5 0 ]
Barcroft and Barron attempted to apply the stimulus
within areas supplied by known sensory nerves. They
found that the area supplied by the maxillary division (V2)
of the trigeminal nerve became sensitive at 34 days, espec-
ially in the region below the eye. Faradic stimulation in
this area caused rotation of the occiput away from the stim-
ulator and contralateral flexion of the neck, a movement
which caused the snout to approach the electrodes. Some-
times movement of the ispilateral forelimb followed or ac-
companied the head movement and, in somewhat older
fetuses, this movement of the forelimb became a constant
component of the response.
By 38 days, faradic stimulation over the infraorbital
foramen caused contralateral rotation of the occiput, fore-
limb extension and extension of the trunk and tail, the
hind limbs moving only passively with the trunk extension.
By 40 days, this series of action components was accom-
panied by opening of the mouth, seen only occasionally at
38 days. At 40 days also, touching a thin glass rod to the tip
of the snout caused head and trunk extension, bilateral
forelimb extension and pelvic flexion, as a result of stimu-
lation of V2 on both sides at once. When only one V
2
area
was so stimulated, however, head rotation to the ipsilateral
side, ipsilateral trunk flexion, extension and partial abduc-
tion of the homolateral, and flexion of the contralateral,
forelimb, with either flexion or extension of the hind limbs,
resulted. These two types of activity were clearly initiated
by the trunk musculature and its maintenance of the re-
sponse posture lasted longer than that of head or limbs.
Various other nerve areas were also tested. Glass-rod
stimulation of the tongue caused only mouth closure at
42 days, but by 59 days evidence of sucking was observed.
[ 5 1 ]
In addition, chewing and swallowing activities were seen
to accompany the sucking movements by 80 days.
The area supplied by the mandibular division (V3 ) of
the trigeminal was markedly retarded over V2 in becoming
sensitive to tactile stimulation. Only on the 50th day did
glass rod stroking in the V3 area cause a response, limited
to opening of the mouth. By 60 days, the mouth opened
and the head turned toward the rod.
The area supplied by the ophthalmic division (V
t
) be-
came sensitive earlier than the V* region, at 40 days. Re-
sponses to stimulation of this area were much the same as
secured from V2.
Other skin areas of the body do not become sensitive
to stimulation with a fine glass rod before the 44th day and
no "generalized" responses were secured elsewhere than in
the facial region.
The studies of Barcroft and Barron, reviewed above
quite incompletely, form one of the most carefully con-
trolled investigations yet made on infrahuman mammals.
They again emphasize that generic differences in the na-
ture of the activity sequence occur, indeed are to be ex-
pected.
The foregoing review of studies on the development of
activity in the infrahuman vertebrate phylogenetic series
demonstrates that two major points of view (and several
minor ones) have been advanced in explanation of the
fundamental nature of the behavioral sequence. One of
the major concepts may be termed the Coghillian, the
other the Windelian, for their respective principal pro-
ponents.
The Coghillian concept maintains that reactive be-
[ 5 2 ]
havior starts as a total pattern, in the sense that all of the
neuromuscular mechanism which is, capable of functioning
participates from the beginning. This total pattern ex-
pands as more and more of the neuromuscular mechanism
becomes functional. A necessary part of this development
to the functional level of nerves and muscles in the normal
cephalocaudal differentiation of the organism is the inte-
gration of the new neural elements with those that have
already become functional. Such integration is a part of the
morphological process of development and is completed
before function of these new elements begins.
When the total pattern has been completed, specific,
more or less local, reflexes ("partial patterns") first appear
("become individuated"). Such specific reflexes also tend
to appear in a generally cephalocaudal and proximodistal
direction. They are made possible by the further develop-
ment of new neural elements which are likewise integrated
into the total pre-existing functioning nervous mechanism
before they can begin to participate in the expanding ac-
tivity of the embryo. It would thus appear that these spe-
cific local reflexes are by no means unitary and simple, but
are derived as a result of the increased complexity of the
nervous system during its development.
The exact nature of activities which compose the de-
veloping behavioral sequence of any given animal form
differs, at least in particulars, from that of any other.
Nevertheless, the broad concept of an expanding total pat-
tern from which specific local activities are secondarily de-
rived seems to hold throughout the vertebrates.
The second major viewpoint, originated by Swenson
but receiving its major support from the conclusions of
Windle, is based on an almost diametrically opposed con-
[ 5 3 ]
cept. Windle has denied that there is any thread of basic
similarity running throughout the vertebrate ontogenetic
series. He states that Coghiirs concept applies only to am-
phibian forms. In the mammals, Windle has claimed that
behavior develops by the appearance of a localized, non-
integrated, unitary response, then, seriatim, of others
which become secondarily integrated with those pre-exist-
ing. In the birds, he believes the activity may be interme-
diate between the total pattern and unitary types of re-
flexes, resulting in an admixture leading preponderantly
toward the latter concept.
In addition, Windle has stated that the mammalian
unit reflexes, described by him, are elicitable only during
an extremely short period of time, measurable in seconds
rather than minutes, and only under the most favorable
conditions. Manipulation of the uterus or of the fetus al-
ters the normal O2-CO2 content of the fetal blood almost
instantly and induces relatively immediate anoxia or as-
phyxia. In consequence, responses secured after the first
moments of exposure are abnormal, in his view, their ab-
normality being due to anoxia. This concept of the method
of origin of overt behavior in the mammals is extended by
Windle to that of man, to be considered in the next chap-
ter. Following consideration of the varying views concern-
ing the development of embryonic and fetal behavior in
man, the two major basic concepts will be further discussed
in the last chapter.
5
Ckapter 2
Tke Sequence in Human Fetal
Activity
I
T HAS PROVED a difficult task to determine the number
of human embryos or fetuses the early activity of which
has been described by scientific observers only, and it is
impossible to give any data regarding the very early state-
ments of those non-scientists who have observed them. In-
asmuch as even the earlier observers with some scientific
affiliations gave only casual notes of quite uncontrolled ex-
aminations, a complete list of references to them can have
only historical interest. Although Erbkam (1837) appears
to be the first scientist to record observations of a living hu-
man fetus, to be followed by Pfliiger (1877), Zuntz
(i877),Rawitz (1879),Preyer (i885),Strassman (1903),
Yanase (1907),Krabbe (1 91 2), Winterstein (1914), Ber-
sot (1920-1921), Woyciechowski (1928), Liesch (1946),
Davis and Potter (1946), Langreder (1949), and Barcroft
(unpublished), adequately controlled studies did not be-
gin to appear until about 1920, with the first publication of
Minkowski.
Minkowski's studies (1920-1946) covered some 75 or
more fetuses and constitute the classic fundamental con-
tribution on the subject. His work on the development of
the plantar reflex (1923,1926) bids fair to retain its valid-
ity for a long time to come. Nevertheless, Minkowski's in-
terpretations were those of a psychiatrist relatively un-
trained in embryology. In consequence, the earlier ac-
tivities observed by him are open to other interpretations.
[ 5 4 ]
The Sequence in Human Fetal Activity
[ 5 5 ]
His observed facts are unexpectedly accurate in view of the
complex movements exhibited by young human fetuses
and the rapidity with which they are executed, for Min-
kowski did not have the advantage of objective motion-pic-
ture recording of the activities observed. To observe,
analyze, and record movements as they occur requires a
master observer and this Minkowski has proved himself to
be.
In 1924, Bolaffio and Artom published the results of
their observations on some 28 human fetuses, insofar as
they can be identified as separate cases. Unlike Minkowski,
these investigators exercised little effort to maintain a nor-
mal condition of the fetuses during their observations and
the methods of stimulation used were not well adapted to
the problem. Their results, therefore, cannot be evaluated
satisfactorily.
These were the only two studies of any magnitude
when the group in the Department of Anatomy at the Uni-
versity of Pittsburgh School of Medicine began, in 1932,
the studies which will be discussed later in this chapter.
However, in 1937, Windle and Fitzgerald published their
first paper on seven human embryos 7 to 8 weeks of age,
none of which moved. Later, Fitzgerald and Windle
(1942) described the activities of four of their total of 17
embryos observed to that time and these findings deserve
detailed presentation.
Two of Fitzgerald and Windle's four active embryos /
were observed while the placental connections and amnion
were intact and within a very short time after exposure by
a uterine incision. These two embryos were, respectively,
22.5 mm. (7J4 weeks) and 23 mm. (stated to be 8 weeks)
in crown-rump (CR) length. In the smaller of these, the
[ 5 6 ]
authors believed that anoxemia had already begun, as no
movements resulted from tapping on the amnion. How-
ever, "strong" stimulation with a needle in the mouth re-
gion caused a contralateral flexion of the trunk. On still
stronger stimulation, the arms and legs moved with the
trunk in the resulting flexion. Only pressure produced re-
sponses, no reflexes being secured by touch or stroking.
Although homolateral flexions of the trunk were more
numerous, contralateral flexions were observed. Pinching
or heavy stroking of the trunk or limbs caused "contrac-
y
) tions of the muscles in the immediate vicinity" (p. 162).
When the amnion of the 23 mm. embryo was tapped
"so lightly that the fetus within was scarcely jarred,"
"quick reactions of the limbs or lower trunk" occurred.
The "arms or legs moved sharply caudad" (p. 162) at times
and once the trunk flexed ventrally. Stronger tapping on
the amnion caused simultaneous movements of all ap-
pendages.
The other two fetuses were 20 mm. and26 mm. in CR
length. In both, the placenta had just separated from the
endometrium. In the 20 mm. embryo (7% weeks), no
reflexes were observed on the operating table, but a minute
after placental separation, while it was in Locke's solution,
"strong" mechanical stimulation in the maxillary area
caused definite contralateral flexion of the trunk, without
participation of either arms or legs.
When the amnion of the 26 mm. CR (8
l
A weeks)
embryo was tapped, the arms and legs responded as they
had in the 23 mm. embryo, but these movements were
accompanied by lateral flexion of the head and trunk.
Fitzgerald and Windle believe that anoxia was developing
[ 5 7 ]
in this specimen, since the embryo was "less excitable"
than the 23 mm. case.
sje | # # # *
The Pittsburgh studies on the physiological responses
of human embryos and fetuses were begun in 1932 by the
writer (1936-1944). Morphological studies on the physi-
ologically tested cases incapable of survival were started al-
most immediately by Dr. Ira D. Hogg and are being con-
tinued by Dr. Tryphena Humphrey. Some of the results of
the morphological work will be presented in the third
chapter, but the remainder of this chapter will be devoted
to the sequence of fetal activity as shown by the physio-
logical studies. However, it seems advisable first to review
the factors which might modify the validity of the results
secured and the measures taken to combat their effects.
In general, it may be said that three types of factors
must be overcome, insofar as it is possible to do so /the pro-
gressive anoxic condition of the embryo, maternal anes-
thesia, and a group of physical factors. The progressive
hypoxia of the embryo is, of course, produced by the
placental separation that occurs within one to two and a
half minutes before the initiation of stimulation. If the
placenta, attached by the umbilical cord to the embryo, is
placed under an atmosphere or more of oxygen in a separ-
ate container, it is possible to improve the fetal heart rate
materially and to prolong the period during which it con-
tinues to beat. Both rate and amplitude of the heart beat
may be altered at will by withholding or supplying oxygen.
However, such administration of oxygen to the placenta
has no discernible effect either upon the nature of the re-
sponses elicitable or upon the length of time during which
they may be secured. This factor, then, must be endured,
[ 5 8 ]
for the present at least, unless one may examine fetuses in
utexo as was done by Fitzgerald and Windle. Just how im-
portant or unimportant it may be during the earliest re-
sponses secured will be discussed later (pp. 102-110). Fe-
tuses of about 2 314 weeks or older may be resuscitated and,
during the period of respiration, which increases with the
age of the fetus, activities are entirely normal, as the
hypoxic state is absent.
The second factor affecting the activity of the fetus is
maternal anesthesia in those cases removed operatively. It
has been our experience in Pittsburgh that the barbiturates
(amytal, monosodium isoamyl, ethyl barbiturate, nem-
butal or pentobarbital sodium) rather completely anesthe-
tize the fetus so that activity is eliminated. Injections of
picrotoxin counteract barbiturate anesthesia, but its use in
embryos and very young fetuses presents difficulties, if not
abnormalities of motion, because of the volume it is neces-
sary to inject. Coramine is another drug favored by some
physicians to overcome anesthesia, but its effect is at least
doubtful.
Other drugs tend to reduce fetal activity, but do not
eliminate it, unless given the mother in massive amounts.
Nevertheless, their use is considered in these studies as
inimical to the observation of normal activity, although
there will be further comment on this point (p. 109). Such
drugs are avertin (tribromoethyl alcohol), ether, nitrous
oxide, and morphine. The first two are more apt to elimin-
ate fetal activity than the others.
Certain drugs, as novocaine and similar compounds,
either locally or as spinal anesthetic, demerol (meperidine
hydrochloride), atropin or ephedrin, when given the
mother, appear to have no affect upon fetal activity. In
[ 5 9 ]
consequence, when fetuses are secured by hysterotomy, we
have gradually learned to use for this study only fetuses
from cases where the attending obstetrician has decided to
use local or spinal novocaine and where preoperative medi-
cation consists only of demerol, atropin, or ephedrine.
The physical factors which might affect the normal na-
ture of the results are: release from intra-uterine pressures,
disappearance of temperature control, and physical in-
sults afforded by operative procedures or premature birth.
Very little can be done about the release from intra-uterine
pressures incident to removal of the embryo or fetus from
the uterus by operation or by spontaneous premature birth.
Actually, there is no difference between the activities of
embryos observed within the intact amniotic sac, where
the pressures differ perhaps little from those in utero, and
after their release from it. Older fetuses gain a freedom of
action of the extremities which is largely denied them
within the confining uterine walls.
Temperature can be controlled as readily outside the
uterus as within, except for the moment of transfer. We
have used a large and carefully regulated constant tempera-
ture bath and have utilized "gold glass" in front of the
photographic spotlights to prevent their heat from affect-
ing the fluid. Embryos and all fetuses too young to respire
are examined in physiological saline, Tyrode's solution, or
some similar isotonic medium, kept at body temperature.
Fetuses which have been resuscitated are examined in
heated "premature beds" in rooms warmed to 85
0
F.
The most serious factor in its effect on the embryo or
fetus is the manipulation of the membranes and fetus in
hysterotomy. Observation of the operator's care in avoid-
ing trauma and examination of the fetus can give a basis
[ 6 0 ]
for estimation of the seriousness of the damage done. How-
ever careful the removal has been, there is some insult.
Nevertheless, the few young fetuses known to have been
damaged, as shown by multiple petechial hemorrhages
under the skin, have often responded actively and in a
manner indistinguishable from that of fetuses not so af-
fected.
Two methods have been used to record the activities
exhibited during the observations: motion pictures and
dictated notes. Chief reliance has been placed upon the
motion picture records, which permit exact and repeated
analysis of the components of each response and are en-
tirely objective in nature. Dictated notes are made to re-
cord activities which it was not certain had been "caught"
by the motion picture camera. Such notes are rarely used
in the study unless supported by the cinematographic
record. Ordinarily, 35 mm. negative film is used for normal
operation, but 16 mm. negative film has been used for
"slow motion" recording and 16 mm. reversible film was
employed during the war when 35 mm. was not available.
All positive prints are made on 16 mm. film.
In the Pittsburgh studies 131 cases have been ob-
served.
6
The fetuses range in age from 6Vi weeks of men-
strual age to a postmature of 45 weeks. It is interesting to
note that four of the 131 cases were twins, but, with one
exception, only one twin of each pair was studied inten-
sively. Of the 131 cases observed, two failed to exhibit
movement probably because they were too young (6V2
weeks), 14 failed to show responses because of deep anes-
thesia or injury, and 31 moved, but their activities were
possibly reduced or modified by the administration of anes-
thetic or hypnotic drugs to the mother. The remaining 84
[ 6 1 ]
were secured by hysterotomy or hysterectomy under novo-
caine spinal or local anesthesia or, in the cases of older
fetuses, were born spontaneously without the use of drugs.
These 84 and the two which were probably too young to
move may be considered entirely unanesthetized. The re-
port is based on the 84.
The standard method of stimulation employed to test
for reflexes is stroking restricted areas of the skin with a
hair. The hairs used are calibrated to provide maximum
pressures of 10, 25, 50, 100 mg, 2 and 5 gm. The 10 mg.
hairs, are human; those capable of exerting greater pressures
are horsehairs of carefully selected diameters. Each hair
is mounted in a glass tube with hard paraffin and calibrated
by the cut-and-test method. The mounting is held ver-
tically over one pan of an accurate chainomatic balance,
with the desired weight on the other side. Each hair is then
cut until it keeps the pan on which it is pressed in exact
balance with the weight when bent maximally but with
only the tip of the hair touching the pan. The tip of the
hair is then beaded with a tiny drop of inert liquid cement
to prevent its cut end from abrading or penetrating the
skin. When this has dried smoothly, the hair is retested to
be certain its "pressure value" remains the same. Repeated
tests have shown that the warm solution used for the em-
bryo does not alter a hair's "pressure value" during the
period of any observation.
As the full pressure of such a hair can only be exerted
when it is forced down upon one point and as such "spot
stimulation" is rarely an effective stimulus, the pressure
exerted in the light stroke over the skin used to produce
exteroceptive stimulation is far less than its "pressure
value." Beaded glass rods for pressure stimulation and elec-
[ 6 2 ]
trie stimuli are also used, at times. Unless otherwise speci-
fied, the results given in the account which follows are
from stimulation with a hair.
The determination of the age of a human embryo or
fetus is very difficult. Many women, and especially those
in a ward population, have only a hazy idea of the
date of onset of the last menstrual period ( L. M. P. ) . Wi th
some notable exceptions, patients will furnish a dateany
datefor the L.M.P., when urged. Examination of the
level of development of the embryo or fetus proves that
little dependence can be placed upon the date given for
that particular L.M.P. In consequence, some other means
of staging embryos and fetuses by age must be used. There
are innumerable formulas and tables for determining age
from length. Only two seem to have value. The tables pub-
lished by the late George L. Streeter in 1920 were used in
these studies, partly because they were the only reliable
ones available when these studies were begun and partly
because experience has indicated that they fit the Pitts-
burgh population well. In 1941, Edith Boyd published for-
mulas and curves for age determination based on both
crown-rump and crown-heel length. Boyd used a wide
range of data, some from Minnesota and some from other
regions. Her tables are probably more scientifically vali-
dated than Streeter's, but we have continued to adhere to
our original staging. Furthermore, we have used men-
strual age throughout, as attempts to "correct" it to "ac-
tual" age merely confound confusion.
* * * * * *
Until about the middle of the 7th week of jaeristrual
age^ the human embryo appears to be incapable of any
type of reflex activity. There is certainly no area of integu-
[ 6 3 ]
merit sensitive to exteroceptive stimulation before this
time. The musculature of the human embryo may be elec-
trically stimulated to contraction beginning with the
latter part of the 6th week, but contractions so produced
are sharply localized at this or at any subsequent age. There
appears to be no true myogenic^cti3idty-iH human embryos.
From approximately the middle of the 7th to just be-
fore the 8th week, when the embryo is between 20 and 23
mm. CR, stimulation by lightly stroking the upper or
lower lip or the alae of the nose with a hair exerting a maxi-
mum bending pressure of 10 or 25 mg. causes typically a
contralateral flexion of the neck and uppermost trunk,
with little or no participation of the upper extremities and
none of any other portion of the body. Two such cases
have been observed in the Pittsburgh studies. From one
(93A, 20.7 mm. CR) , only a single response without vis-
ible participation of the upper extremities was elicited.
From the other (No. 1 31 , 22.6 mm. CR) , a large number
of identical responses was secured. In these the upper ex-
tremities quivered slightly during the contralateral neck
and upper trunk flexion, as though its musculature were
attempting to contract. This quivering motion extended
even to the fingers, but the upper extremities did not move
otherwise. The skin area sensitive to stimulation was
limited to the region immediately about the mouth. Al-
though previously it was believed that some slightly older
embryos ( 8+ to 854 weeks) showed sensitivity in the area
supplied by the maxillary division of the fifth cranial nerve
a shade earlier than in that supplied by the mandibular di-
vision, this was not confirmed in either of the embryos here
described.
To these two cases from the Pittsburgh series may be
[64]
added the two, previously described, which were observed
by Fitzgerald and Windle (1942). Their 20 mm. embryo,
showing no responses on the operating table, gave definite
contralateral flexions of the "trunk," without arm move-
ment, following "strong" mechanical stimulation of the
maxillary region. Their 22.5 mm. embryo, also showing
no movements on tapping of the amnion, exhibited both
ipsilateral and contralateral flexions of the trunk, both
arms and legs moving with the trunk, on stimulation with
a needle in the mouth region. The latter embryo responded
more like one of 8
l
A weeks and was, in fact, estimated by
them at 8 weeks of age.
There is no active or unanesthetized embryo of ex-
actly 8 weeks in the Pittsburgh series, but there are six em-
bryos of 8 V2 weeks (25 to 28 mm. CR) which moved, one
of which is subject to suspicion of being possibly slightly
anesthetized because morphine had been administered to
the mother. The five unanesthetized cases all responded to
unilateral exteroceptive stimulation in the mouth region
by light stroking with a hair, and the responses were, in
each case, practically identical with every other when re-
peated. They were, thus, stereotyped.
The typical response at 8% weeks consists of the fol-
lowing elements: 1) chiefly contralateral flexion of the
neck and trunk, though two ipsilateral flexions are re-
corded; 2) extension (backward movement) of the arms
at the shoulder, without participation of the elbow, wrist,
or fingers; 3) rotation of the pelvis toward the contralateral
side, with no noticeable independent participation of the
lower extremities. The neck-trunk flexion and the arm ex-
tension were simultaneous. The pelvic (rump) rotation,
slight in the younger embryos, occurs with the trunk flex-
[ 6 5 ]
ion in these younger cases but tends to have an extremely
short period of delay in those of increasing length (and,
presumably, older) when the rotation has become some-
what more pronounced.
The relaxation of the movement (return to posture) is
accomplished first by the arms, then by the neck, trunk,
and rump, which tend to hold their flexion-rotation pos-
ture momentarily. Because the body of the embryo rolls
as a result of the trunk flexion, return to the exact position
occupied before the response does not always occur. How-
ever, the difference is slight. A series of such responses may
be elicited at short intervals for as long as three to four
minutes after beginning placental separation. Each re-
sponse is, within relatively minute biological variations,
identical with every other, hence these responses are also
stereotyped. However, as the time interval between the re-
sponse and beginning placental separation lengthens and
the degree of anoxia increases, the threshold rises gradually,
requiring increasingly higher "pressure value" hairs to
evoke a response.
As the young fetus approaches the g^-week age, cen-
tering around 34 mm. CR, the rump rotation becomes still
more marked and tends to become separated from the
usually contralateral trunk flexion and bilateral arm exten-
sion at the shoulder as a distinct phase of the response. Al-
though the vigor of movement of the caudal end of the
trunk has increased over that seen at 8*/4 weeks, the skin
area sensitive to stimulation remains the same. Further-
more, spontaneously executed movements in the form of
the typical response first to one side, then to the other, are
clearly evident at 9 to 9 Yi weeks, but only for a very short
[ 6 6 ]
time, limited to about one minute after the beginning of
placental separation.
There is some evidence that such spontaneously exe-
cuted activity might be present at 8
l
/z weeks. One 854-
week embryo (No. 116, 27.1 mm. CR) reacted repeatedly
to stimulation in the sensitive area by side-to-side repeti-
tions of the typical response. Another (No. 24, 26 mm.
CR) gave a single contralateral-ipsilateral response to peri-
oral stimulation a few minutes after delivery. These in-
stances were in response to definite exteroceptive stim-
ulation, but were of the bilateral nature typical of spon-
taneous activities. The similar, but truly spontaneous,
movements seen at 9 to 9 J4 weeks were not in response to
identifiable external stimulation.
At 9 Vi weeks, another method of exciting the typical
response, consisting of lateral trunk flexion, bilateral arm
extension, and rump rotation, was accidentally discovered.
To center a fetus (No. 16) under the lens, it was rapidly
rotated by the umbilical attachment of the membranes
without any contact being made with the usual reflexo-
genic perioral area. A very complete typical response re-
sulted. This was formerly (1944 et ante) interpreted as
a vestibular response. It is absent in two 8^2-week fetuses
which were similarly rotated. So far as may be determined
histologically, the vestibular apparatus appears complete
at 9V2 weeks. However, there is much evidence that vesti-
bular function does not appear in cat or sheep until much
later. Windle and Fish (1932), on the basis of their ro-
tatory experiments with cat fetuses and ablation of the
vestibular apparatus, present evidence that the vestibular
righting reflex in this form does not appear earlier than the
54th day of gestation, although body-righting reflexes ap-
[ 6 7 ]
pear on the 50th day. Barcroft and Barron (1937, 1939b)
state that vestibular responses do not appear in the sheep
until the 43rd day after birth, although the "neck righting
reflex," as they term it, can be evoked by the 50th day of
gestation. In consequence, the post-rotation response in
the human fetus would appear to be a body-righting reflex
caused by the disturbance of head-body relations. How-
ever, Minkowski (1936) states that both "cervical tonic"
and "labyrinthine" reflexes appear in the human fetus dur-
ing the fourth month. This age is, of course, far greater
than that of the 9 Yi -week fetus in which the response un-
der discussion was found.
At the 9 Yz -week level, passive extension of the fingers
has caused flexion at the wrist, elbow, and shoulder. These
appear to be typical stretch responses, proprioceptive in
nature.
If the description so far given of responses from about
jYi to 9Y2 weeks has accomplished its purpose, it should
be evident that the earliest responses secured were flexion,
predominantly contralateral, of neck and upper trunk
without participation of the extremities. This was followed
by more complete trunk flexion with extension of both
upper extremities at the shoulder-joint and a slight pelvic
or rump rotation, without movement of the lower extremi-
ties as such. Next in turn came marked lateral flexion of
the trunk, preponderantly contralateral, bilateral arm ex-
tension, and such complete rump rotation that the lower
extremities appeared to participate, especially so in the
body-righting response. Except for a short period around
8 weeks, every step in the caudal progression of the re-
sponse has been observed. During the period from jYi to
about 10 weeks, the only skin area sensitive to a light strok-
[ 6 8 ]
ing with a hair is limited to the perioral distribution of the
maxillary and mandibular divisions of the fifth cranial
nerve. These early responses would appear to constitute a
continually expanding total pattern. No other explanation
fits the observed phenomena.
From <)
l
A to 10V2 weeks there is no change in the na-
ture of the response to stimulation in the perioral region.
At about IOVI weeks the palms of the hands become sensi-
tive to strokes with a light hair, partial closure of the fin-
gers resulting. In the finger-closing response, the thumb
frequently does not participate. Three to five days later,
stimulation of the volar or lateral surfaces of the upper arm
may cause independent arm extension at the shoulder; of
the upper eyelid, contraction of the orbicularis oculi mus-
cle; and of the sole of the foot, plantar flexion of the toes.
These responses have so far appeared in the order given,
although independent arm extension has only rarely been
seen. It will become apparent that it is easier to evoke re-
sponses from fingers or toes than from brachia, forearms,
thighs, or legs.
At about 1 1 weeks, the previously typical response to
perioral stimulation becomes modified by the appearance
of an element of trunk extension, either alternating with
or partially combined with the lateral flexion. Arm exten-
sion, which has hitherto accompanied the trunk flexion, is
giving way to medial rotation of the brachia, which will
reach its full development in another few days. There is
also some slight rotation of the thighs in the total pattern
response. Finger closure, still only partial and seldom ac-
companied by thumb flexion, follows palmar stimulation.
Stimulation of the skin of the forearm may occasionally
cause forearm pronation. Plantar flexion of the toes, with
[ 6 9 ]
or without participation of the hallux, follows stimulation
of the sole of the foot. Contraction of the orbicularis
oculi and of the corrugator muscles may be elicited by
stimulation either of the upper eyelid or of the eyebrow
region of the forehead.
It will be noticed that the skin area sensitive to stimuli
has been expanding. No longer limited to the perioral re-
gion, it now includes all skin over the front of the face and
the whole of the upper and lower extremities. However,
stimulation of the region innervated by the maxillary and
mandibular divisions of V gives rise to total pattern type
activity, of other areas to local responses only. On the
trunk, there is no sensitive skin evident a t n weeks. Only
light stimuli are necessary to evoke responses over the face,
hands, and feet, and the fetuses are active in responding.
The total pattern type of response is thus modified by the
appearance of a number of specific responses.
At n Vi weeks, the reflexogenous zone (area sensitive
to stimuli) gives evidence of spreading downward over the
upper chest. At this age, activities already present have be-
come further perfected and additional new ones appear.
Extension of the trunk is increasingly, but by no means en-
tirely, replacing the hitherto typical lateral flexion. Trunk
extension is especially marked when stimulation is in the
midline of the face, but also occurs on stimulation else-
where on the face. As one approaches closer to the end of
the period during which reflexes can be elicited from an in-
dividual specimen, the lateral flexion type of response
tends to increase in incidence, probably because of in-
creasing asphyxia, although the locus of stimulation plays
an important role. At this age, the medial rotation
of the brachia accompanying the trunk movement has
[ 7 0 ]
increased to the point where the two hands approach
one another, although only the fingers touch. Rotation of
the face to the opposite side has appeared in response to
stimulation of the side of the face, but midline facial stim-
ulation causes head extension. Head rotation is usually ac-
companied by lateral trunk flexion, head extension by
trunk extension, and it is with the latter type of response
that the medial rotation of the brachia reaches its maxi-
mum. Arm abduction or wrist or forearm flexion may fol-
low stimulation of various areas of the skin of the upper
extremity. Stimulation of the sole of the foot causes plan-
tar flexion of the toes, flexion of the knee and thigh, and
then a quick knee extension, resulting in a kick.
By 12 to 12V2 weeks, the reflexogenous areas of the skin
have not appreciably increased, but unilateral facial stimu-
lation in the perioral region causes marked extension of
neck and trunk accompanied by a wide variety of move-
ments at many of the joints in the upper extremities. The
responses are still quite mechanical in their execution and
the fetuses continue to be very active, but there is some-
what greater variety in the exact components of the re-
sponses, one differing from another in the accompanying
movements. The general response may be evoked by stimu-
lation of either the region supplied by the maxillary divi-
sion of the fifth cranial nerve (V2) or that to which the
mandibular division ( V
3
) passes. On one occasion, when
V
3
had been stimulated, a contraction of the orbicularis
oculi muscle of the same side was observed. Perioral stimu-
lation has also caused, as activities accompanying the total
pattern, such movements as: rapid contralateral elbow
extension, then its flexion; lateral, then medial rotation of
contralateral arm; and elevation of the shoulders.
[ 7 1 ]
At this age there are several specific reflexes which may
be evoked. In at least two cases, stimulation over one eye-
lid was accompanied by a downward rotation of the eye-
balls, made evident by the shift in the highlights on the
still fused lids. Lip closure follows stimulation of the lips
and, when repeated, may result in swallowing. Digital
closure is incomplete but is usually accompanied by wrist
flexion or a combined flexion-abduction motion. In
week specimens, stimulation of the sole of the foot may
result in dorsiflexion of the hallux and toe-fanning, or in
plantar flexion of hallux and toes, though flexion of the
hallux is likely to be less complete than that of the other
digits, if not entirely absent.
By i2
1
/i weeks, in the majority of cases, the total pat-
tern activity is rapidly fading, being replaced by an in-
creasing number of specific reflexes. Coghill, himself, has
called these "partial patterns." Within the next ten days,
the total pattern has almost completely disappeared, to be
evoked only by extremes of stimulation. Some human
fetuses have reached this stage by 13 weeks of menstrual
age. Others at presumably the same age level still exhibit a
stereotyped total pattern. All below the age of 13 Vi to 14
weeks tend to return to a total pattern type of response be-
fore asphyxia and other causes blot out further activity.
Specifically, at 13 weeks, unilateral stimulation in the
perioral region may result in any one of the following re-
sponse types, in addition to a total pattern shown by a few:
1 ) ipsilateral arm extension, only; 2) flexion of the head,
bilateral arm extension, and extension of the ipsilateral
thigh; 3) head extension, contralateral arm extension, and
ipsilateral thigh flexion, knee extension, and dorsiflexion of
the foot; 4) trunk and neck extension, rotating into a con-
[ 7 2 ]
tralateral trunk flexion, without extremity participation;
5) contralateral turning of the face and flexion of the head
away from the stimulus; or 6) turning of the face toward
the stimulus. Stimulation over the extremities with hairs
having pressure values of from 25 to 100 mg. or of 2 gm.,
depending on the degree of anoxia, may result in purely
local responses of the part touched or of the entire extrem-
ity. This incomplete listing of responses exhibited serves
to illustrate the transition from total pattern with some
specific reflexes to a manifestation of specific responses al-
most solely, found between 13 and 14 weeks.
By 13 Vi to 14 weeks, except for the back and top of the
head, almost the entire surface of the body is sensitive to
stroking with a light hair. As noted, the total pattern type
of response has practically disappeared. Unilateral stimu-
lation of the face causes contralateral turning of the face
alone, an avoiding response, accompanied by elevation of
the angle of the mouth and nasal ala on the same side, if
the hair touches the side, of the nose, or a contraction of the
orbicularis oculi muscle, if the stimulus is applied over the
eyelids. If the lips only are stimulated, they are pressed to-
gether, without head rotation. Touching the inside of the
lips or mouth may cause tongue movements. Swallowing
sometimes follows lip stimulation, but usually only on
repetition of the stimulus. Finger closure is complete, the
pollex rarely entering into the formation of the resulting
"fist," but the closure is relatively momentary, there being
no sustained grasping response as yet. The quality of the
responses is altered. Instead of the mechanical, stereotyped
movement seen earlier, the various activities are graceful
and "flowing." The fetus is very active.
The 13 J4- to 14-week period is an important one in the
[ 7 3 ]
development of human fetal activity. During the period of
6 to 6
L
A weeks after exteroceptive responses began as a
total pattern at about 7
L
A weeks, many of the specific re-
flexes of the neonate have appeared, replacing the total
pattern which has practically ceased to exist as such. Those
reflexes which have made their appearance are concerned
primarily with trunk, extremities, head, and face. They
are not all in the final form they will assume, but, with the
addition of a number not yet present, they lay the frame-
work for gradual development into the reflexes of postnatal
life. Among those reflexes which have not as yet appeared
are, primarily, grasp, respiration, and phonation (thelatter
being of course dependent on the former), the suctorial
reflex, and the tendon reflexes. Furthermore, the fetal mus-
cles are weak and ineffective. In the remaining weeks of the
gestation period, the increase in volume and strength of
the muscles, the development oadditional neuromuscular
mechanism, and the growth of the lungs will occur.
During the remaining period of gestation, there will
be almost no change in the sensitive areas of the body, as
the back and top of the head remain relatively incapable of
evoking reflexes for some time to come. There will, how-
ever, be a marked maturation of already existing reflexes
and certain ones, now absent and probably requiring new
central and neuromuscular connections, will gradually
emerge.
By 15 weeks, there is definite evidence of maintained
finger closure, sometimes, though rarely, noticeable several
days earlier. This is the first appearance of grasping. At this
time also the abdominal muscles are capable of sufficient
contraction to provide a feeble sort of abdominal reflex
when the stimulating hair is drawn across them. The fetus
[74]
is active and the pre-existing reflexes are already becoming
more vigorous as muscle maturation proceeds.
About a week later, the fetus begins to undergo a pro-
found change in its reactivity. Periods of sluggishness tend
to alternate with bursts of activity and the threshold for
stimulation has increased. As age progresses, the periods of
relative inactivity of the fetus become more prolonged and
the bursts of activity more difficult to evoke. This is char-
acteristic of the period beginning at about 17 to 18 weeks
and continuing until the fetus may be caused to breathe at
about 23% weeks. Barcroft and Barron (1937) observed
the same thing in sheep fetuses with intact placental con-
nections. The cause is still undetermined. In the human
fetuses observed, in which the placental connections had
been interrupted, it is possible that the sluggishness may
be due to accumulations of CO2 which are eliminated by
the ventilation of the lungs on resuscitation. Certain it is
that the establishment of respiration soon abolishes any
such sluggishness.
By 17 weeks, stimulation of the upper lip causes its
protrusion. Similar gentle stroking of the lower lip at this
age produces lip closure, as was the case with either lip at
i2
L
A weeks.
Gasping, as a terminal agonal phenomenon without
external stimulation, has been observed in fetuses as young
as 13 weeks. This is accompanied by chest contractions,
indicative of respiratory attempts, albeit ineffectual in
causing breathing. By i8
L
A weeks, weak chest contrac-
tions, later to be accompanied by abdominal contractions
and expansions, the latter proving diaphragmatic activity,
have been observed in some cases in response to chest stim-
ulation, or spontaneously, though these also are ineffectual
[ 7 5 ]
in causing respiration. The earliest Pittsburgh case in
which effective respiration, self-sustained for a brief period
when once initiated, was observed was 23V2 weeks old, as
just noted. At i%Vi weeks also, weakly effective grasping
is present. A thin, moist glass rod laid in the palm will be
held by maintained finger closure, so that it is possible to
draw the arm across the chest by pulling the stimulator in
the direction of its long axis. The thumb is not constantly
involved in this primitive grasping. The fetus is notably
sluggish in responding at this age. However, when reflexes
are elicited by 100 mg. or 2 gm. hairs, the activity is vigor-
ous.
By 20 weeks, stimulation of either upper or lower lip
causes its protrusion and by 22 weeks there is definite purs-
ing of the lips as well as protrusion. This is a preliminary
step in the development of the suctorial response, but no
actual suction is exerted as yet.
Between 18 54 and 23% weeks, the chest and abdomi-
nal contractions and expansions increase in incidence and
amplitude, until brief but effective respiration, especially
after use of the respirator, occurs. It may be that some
fetuses younger than 235/2 weeks might be resuscitated, but
so far it has not been possible in this series. Whenever
respiration occurs, phonation-only a high-pitched cry, of
courseaccompanies its initiation and may be repeated at
intervals following the establishment of breathing. The
establishment of respiration for 20 minutes or more com-
pletely abolishes the sluggish nature of the fetus. As its
color begins to improve, for most fetuses of this age have at
first a slightly cyanotic appearance even with intact pla-
cental connections, the fetus becomes more active than at
any time heretofore. The arms and legs are very actively
[ 7 6 ]
moved in all direction, the trunk is flexed and extended in
all planes and it becomes possible to elicit various tendon
reflexesankle clonus, knee jerk, etc.not hitherto ob-
tainable. The change in activity following respiration is
startling, although, except for the possibility of eliciting
tendon reflexes, no new activities appear. Indeed, little new
appears during the next two or three weeks. It is almost as
though the fetus were developing its respiratory capacity at
the expense of the appearance of other new responses. This
may, at least in part, be the result of the explosive appear-
ance of "spontaneous" activity which renders the eliciting
of reflexes difficult of evaluation. As a rule, it is almost im-
possible to determine whether a movement following
stimulation was caused thereby or would have occurred
without it.
By 25 weeks, all previous reflexes have greatly matured
and the muscular system is increasing in strength. Self-sus-
tained respiration, once begun, may last for over 24 hours.
Tendon reflexes have become stronger and more com-
plete. The eyelids may be opened and closed without stim-
ulation by the observer and conjugate movements of the
eyes, especially in the lateral plane, are evident. At this age,
also, a Moro reflex was observed quite inadvertently. It
may, therefore, be elicitable earlier. One fetus, lifted gently
from the premature bed, in which it was being photo-
graphed, in order to change the underlying sheet which
had been discolored by urination, was quickly lowered
again. The almost clonic extension of arms and legs which
resulted were typical of this "startle" reflex, which appears
to have a vestibular origin.
Some obstetricians are of the opinion that a few 25-
week premature infants may be viable. This is conceivably
[ 77]
possible, although the youngest viable fetus observed in
this series was 27 weeks of menstrual age. As previously
noted (p. 62) all ages assigned in this study are menstrual
ages. If by 25 weeks the so-called "corrected actual" age
is intended, it would agree with 27 weeks of menstrual age.
If 25 weeks of menstrual age is intendedand too often
the exact kind of age used is not specifiedit might still
be possible, but not too probable in the light of the Pitts-
burgh observations. Possible viability at 27 weeks of men-
strual age seems to agree with the majority opinion of ob-
stetricians.
The 27-week fetus has greatly matured. The grasp of
one was sufficiently strong to maintain with one hand prac-
tically its entire weight for a short period. In all examined
at this or later ages the grasp is effective and the thumb par-
ticipates most of the time, though feebly.
Little in the way of new responses appears from 27
weeks to birth, either at the usual 40 weeks or in such post-
matures as have been observed. The suctorial response be-
comes matured into an active and audible sucking by 29
weeks. A definite cremasteric reflex has been observed by
32 weeks. At 33 weeks, the tongue may "search" the lower
lip when that organ is lightly touched. For more details of
the activities of the circumnatal period the reader is refer-
red to the work of Arnold Gesell (1945) The number of
fetuses observed in the Pittsburgh studies in this age range
has been inadequate for final conclusions.
Let us now turn our attention to a summary and dis-
cussion of the appearance and maturation of specific activi-
ties. First, however, it seems wise to review the spread of
skin sensitivity to tactile stimulation.
[ 7 8 ]
At 7 Y
2
weeks, the sensitive area of the skin is closely
limited to the area about the lips and alae of the nose, but
by 8 through 9V2 weeks the perioral region of sensitivity
has widened slightly to include the chin and lateral parts
about t he mouth and nose. The facial area of sensitivity
increases upward by loVi weeks to include the eyelids in
some and this region has become sensitive in most by 1 1
weeks. Two or three days after this (11 Vi weeks) the en-
tire face is sensitive in all fetuses.
These are the areas supplied by the trigeminal nerve.
In man i t has been impossible to distinguish with certainty
whether the skin area supplied by the maxillary or the
mandibular divisions (V2 or V3) of this nerve becomes
sensitive first. As already noted, Barcroft and Barron
(1939a ) stated that in the sheep the skin area supplied by
V2 becomes sensitive first. Indeed, in the sheep, Va does
not mediate widespread responses, but Vi does. In the
Pittsburgh studies it was thought earlier that the V2 area
might precede the V3 in becoming sensitive to stimulation
in human embryos, but later cases have indicated that such
an assumption is unwarranted. The area on the sides of
the face toward the ears does not become sensitive much
before 1 4 to 15 weeks, and the back and top of the head ap-
parently remain insensitive until birth.
The palms of the hands become sensitive to tactile
t stimulation in a few cases early in the 10th week, ap-
parently very slightly preceding the appearance of sensi-
tivity in the skin of the arms and forearms, and of that in
the soles of the feet, also found in a few specimens only. By
1 1 weeks all fetuses observed exhibited responses to palmar
stimulation and most to stimulation over the arms and
forearms. Although only some showed evidence of sensi-
[791
tivity of the sole of the foot at this age, all did so by 12
weeks. Sensitivity of the skin of the thighs and legs ap-
peared slightly later than on the soles of the feet.
In response to appropriate stimulation, the first move-
ment executed by the embryo is a contralateral flexion of
the neck at about 7V2 weeks. By about 8 weeks the upper
trunk is included in the usually contralateral flexion and
the upper extremities quiver as if attempting to participate,
but are not otherwise involved in the response. From
shortly after 8 weeks to 8K weeks the typical response is a
contralateral flexion of the neck and an increasingly caudal-
ward-extending portion of the trunk, accompanied by ex-
tension of both arms at the shoulder, and some pelvic ro-
tation. Between 8V2 and 9 Vi weeks, the contralateral flex-
ion of the trunk extends still farther caudally, and the hips
exhibit a progressively more marked rotation toward the
side of the trunk flexion. The arm extension at the shoulder
does not alter its character in this period.
The caudally expanding response just outlined is
evoked by unilateral stimulation with a 10 mg. or heavier
hair stroked in the perioral region supplied by the maxillary
and mandibular divisions of V. It may also be evoked at
g
l
A weeks by rotation of the embryo. From 7V2 to at least
10 weeks of age, this is the only skin area of the body sensi-
tive to such light stimulation. The use of heavy (stiff) hairs
may cause local contractions of the longitudinal trunk
muscles due to direct mechanical stimulation of the mus-
cle tissue. Proprioceptive responses of the bineuronal
stretch type may be elicited from forearm or brachial mus-
cles by gYi weeks. Evidence is lacking in these studies as
to whether stretch reflexes may be elicited earlier, but
Fitzgerald and Windle (1942) report responses of this
[ 8 0 ]
type at j
l
A weeks. The response given to unilateral perioral
stimulation with a light hair is patterned and stereotyped.
It is typically contralateral, but may more rarely be ipsi-
lateral.
By io
l
A weeks, trunk extension begins to replace the
usual contralateral trunk flexion and by 1 1 to 12 weeks, ex-
tension has become the primary type of response to uni-
lateral stimulation in the perioral area. This type of ac-
tivity is usually accompanied by more or less marked
medial rotatioq of the brachia. Neck rotation, turning the
face away from the source of stimulation, occurs occasion-
ally by 1 1 weeks and becomes more evident by 12 weeks
and thereafter. By 13
l
A to 14 weeks, the trunk movements
, .generally cease to be patterned and local activities of many
kinds replace the larger responses.
Contractions limited to the chest make their appear-
ance in the 18th week. This is possibly the age at which
true respiratory movements first appear in human fetuses
in good condition. Gasping, with agonal chest contrac-
tions, has been seen as early as 13 weeks in dying fetuses,
but they have never been seen in cases of that early age
when oxygen was being supplied to the placenta. However,
Windle, Dragstedt, Murray, and Greene (1938) report
respiratory movements of the chest in a fetus of 62 mm.
CR (12 weeks). Chest contractions in the 18th week are
not agonal in character and occur whether or not oxygen
is being supplied to the placenta. These contractions are
slight and occur over the more caudal half of the thorax.
They may be elicited by gently stroking the chest with a
2 gm. hair.
By 20 weeks, these contractions have increased in am-
plitude and at 22 weeks are accompanied by expansions
[ 8 1 ]
and contractions of the abdomen. Such abdominal expan-
sions are unquestionably caused by activity of the dia-
phragm, the contractions of which compress the abdomi-
nal contents caudalward, as in adult respiration. Such ac-
tivity, even at 22 weeks, is ineffectual. It has not been pos-
sible to resuscitate a fetus, even temporarily, before 23V2
weeks, although examination of a greater number of cases
might have shown earlier incidence of self-sustained
respiration.
The time of beginning effective respiration appears to
be dependent upon the level of development of the lungs,
especially of its elastic tissue, which becomes differentiated
at about 28 weeks (Hewer, 1938). When abdominal move-
ments begin, the volume of the chest cavity is little, if at
all, increased. When the diaphragm contracts, the lower
thorax is drawn inward to create a relatively deep trans-
verse depression over the lower rib-cage, deepest in the re-
gion of the xiphoid cartilage. This drawing inward of the
lower ventral thoracic wall appears to compensate for the
increase in vertical diameter of the thoracic cavity caused
by the diaphragmatic contraction. At the time when sus-
tained respiration becomes possible, the depth of the de-
pression begins to diminish and, although it persists in
some fetuses for many weeks, continues to become less evi-
dent. The last evidence of the depression lies over the xi-
phoid cartilage.
Originally regarded as a sign inimical to a good prog-
nosis for survival, it has since been proved that older pre-
mature infants exhibiting some depression may nonethe-
less continue to live and thrive. Belief that the depression
of the chest is caused by inadequate development of the
lungs so that they are as yet incapable of filling the ex-
[ 8 2 ]
panded thoracic cavity is based largely on assumption, but
is strengthened by attempts to inflate the fresh lung.
Though inadequate to prove the point, these inflation at-
tempts indicate a relatively rapid increase in lung volume
between 18 and 25 weeks. The gradual disappearance of
the ventral depression during this period of growth further
increases the possibility that the level of lung growth is
largely responsible for viability of the premature infant.
At 23% weeks, respiration, self-sustained for 2
l
/z hours
after resuscitation, was first accomplished. As attempts at
resuscitation were carried to older ages, it was found that
the duration of self-sustained respiration increased with
age, as might be expected. At 24% weeks, the duration of
such respiration was ^Vi hours; at 25 weeks, 26
l
A hours;
and at 27 weeks, one premature continued to live.
Although the facial area is the only region sensitive to
light tactile stimulation during the early reactions of the
embryo, the facial muscles do not participate in responses
of other components of the body and do not themselves
exhibit movements before 10 to 10V2 weeks. At that age
contraction of the orbicularis oculi muscle may occasion-
ally be elicited in response to stimulation of the upper lid.
By 1 1 weeks, the corrugator muscle often participates in
the response of the orbicularis, causing scowling as well as
"squinting" of the eyelids. These occasional responses be-
come quite constant between 12 and 13 weeks of age when
an upper eyelid is stimulated. Eyelid "squinting" has
once been observed following lower lip stimulation at
13 weeks and upper lip stimulation at 14 weeks. At 13 to
14 weeks, unilateral stimulation near the nose or upper lip
causes an ipsilateral contraction of the quadratus labii
superioris muscle, producing a sneering expression.
[ 8 3 ]
The eyes themselves are difficult to observe before the
lids separate. However, what appeared to be conjugate de-
pression of both eyes was observed through the closed lids
in two fetuses of 12 J4 weeks of age. Beginning at 25 weeks
conjugate lateral movements of the eyes have been
seen when the lids were open, and at older ages conjugate
eye movements in all directions have been observed.
Momentary lip closure may be elicited as early as 12 54
weeks when the rima oris is touched and repetition of such
stimulation may evoke swallowing at this age. Within 2
weeks deglutition becomes an almost constant finding on
repeated lip stimulation. By 13 weeks, light touch on
either lip causes its firm closure, maintained for an appreci-
able period of time. Upper lip stimulation at the rima oris
evokes its protrusion at about 17 weeks, but a similar re-
sponse of the lower lip has not been exhibited before 20
weeks. At 22 weeks, simultaneous protrusion of both lips
results in pursing of the mouth. Definite, indeed audible,
suction has been observed at 29 weeks, but may occur
earlier.
The upper extremities slightly precede the lower in
participation in trunk responses and in exhibiting specific
reflexes. Just after 8 weeks, both brachia extend at the
shoulder as part of the trunk response to unilateral stimu-
lation about the mouth. The quivering of the upper ex-
tremity, including the fingers, observed in one case just be-
fore 8 weeks is almost certainly preliminary to the shoulder
extension. At 10 54 weeks, independent extension of one
arm at the shoulder has been observed on stimulation of
the shoulder. This response is rare, however, and occurred
in a fetus in which partial finger closure had been elicited.
Partial finger closure on light palmar stimulation is a con-
[ 8 4 ]
stant response by n weeks, but may occur occasionally 3
to 5 days earlier. In these early cases of finger closure on
palmar stimulation, medial rotation of the brachia, flexion
of the elbow and forearm pronation have also been ob-
served, but these likewise are rare findings. By 1154 weeks,
medial rotation of the arms at the shoulder is a constant
finding on perioral stimulation. Abduction of the arm at
the shoulder in response to lateral brachial stimulation has
been observed once at this age.
Coghill noted in Amblystoma, as has Youngstrom
(1938) in Rana, Bufo, Acris, and other anurans, that spe-
cific limb responses appear in a proximodistal direction.
Angulo (1932b) has noted that the same principle holds
for the rat. Barron (1950) has called attention to the ear-
lier ontogenetic development of both structure and func-
tion in the limbs of mammals in contrast to those of am-
phibians. This fact appears correlated with the earlier im-
portance of the limbs in the former class of animals. Inso-
far as the Pittsburgh studies afford evidence, the human
fetus appears to be further modified than the rat and other
similar mammals in that, as noted, the strict proximodistal
appearance of limb reflexes is only partially true. It may be
that a greater number of sensory endings per unit area in
the hand of man has led to the earlier development of its
reflexes. '
Halverson (1937, 1943) has made a careful study of
prehension in infants and very young children. He notes
that there are two successive steps in its development. The
first of these is finger closure, the second, grasp. It is signif-
icant that in fetal life the same two steps are evident in the
appearance of reflexes of the fingers (Hooker, 1938). In
this and many other activities, postnatally developed vol-
[ 8 5 ]
untary acts appear in the same sequence in which they are
elicitable as reflexes during fetal life. Attention might
be called here to the unquestioned fact that the order of
postnatal appearance of function on a voluntary basis must
mirror the order of appearance of the antecedent fetal
reflexes.
Finger closure, only partial and usually without the
participation of the pollex, first appears between io54 and
1 1 weeks on stimulation of the palm. It is soon accompa-
nied by wrist flexion. Closure of the fingers becomes com-
plete between 13 54 and 14 weeks, at about which time
thumb opposition has been observed. By 14 weeks there is
a tendency for finger closure to be maintained for a brief
period, but anything resembling true grasp does not appear
for another week and becomes effective only by about 1 8 54
weeks. By 27 weeks, however, the grasp may be sufficiently
strong to support almost the entire weight of the body
with one hand. It has proved practically impossible to in-
duce a fetus to grasp with both hands at the same time un-
til the circumnatal period.
In the lower extremity, aside from the rump rotation
which, like the arm extension at the shoulder, is part of the
trunk response, the toes respond to stimulation of the sole
a little before responses of the thighs and legs may bp
elicited. Occasionally, plantar flexion of the toes may be
elicited in more advanced fetuses a little over 10 54 weeks
of age. By 1154 weeks, the toes constantly respond to sole
stimulation. The response may be either plantar flexion or
dorsiflexion. Some fetuses appear to exhibit one or the
other with greater facility, but almost all show both types
of response on successive stimulation. At 1154 weeks, toe
responses may be accompanied by flexion of the knee and
[ 8 6 ]
at first flexion, then extension at the hip, resulting in a
kick. Within a week (12 to 12V2 weeks) and thereafter,
response to sole stimulation consists principally of dorsi-
flexion of the hallux, with toe fanning, dorsiflexion of the
foot at the ankle, and flexion at knee and hip.
s{e # sjc # # #
These, then, are the principal results of the Pittsburgh
studies on the physiological responses of human embryos
and fetuses. There is much yet to be done. Many aspects of
reactive behavior have not even been attempted as yet,
e.g., no studies on the functioning of the organs of special
sense have been made. In other regards, only an inconclu-
sive beginning has been made. It is hoped that progress will
be made along these lines as the studies are continued.
Ckapt er 3
Tke Significance of Structural and
Functional Interrelationships
in Prenatal Activity
I
N THE TWO preceding chapters, the more significant
contributions on vertebrate embryonic and fetal overt
behavior have been reviewed and the current results of
an intensive study of human fetal activity have been pre-
sented. The review has indicated that various investigators,
even in studies on one vertebrate class, have secured widely
varying data. This is perhaps to be expected, as so many
variables exist, and too many students of this complex prob-
lem have fastened their attention upon this or that aspect
to the exclusion of others. Before the whole question of the
development of activity throughout the vertebrate series
will be solved, additional carefully planned and controlled
observations must be made.
In the foregoing accounts, attention has been given to
overt behavior alone. Very nearly the only indication that
this externally visible activity is dependent upon the in-
ternal organization of the developing embryo, larva, or
fetus has been the often repeated mention of the neuro-
muscular mechanism of the organism. Any living thing
exists in an external environment. However, every living
thing has an internal organization, which in most inverte-
brates and all vertebrates takes the form of organ systems.
The interaction of the activities of these organ systems
upon one another constitutes the internal environment of
[ 8 7 ]
[ 8 8 ]
the whole animal. Furthermore, each organ system exists
in an environment composed in part of elements of the ex-
ternal environment of the whole creature and in part of
the elements furnished by the activities of other interact-
ing organ systems. This is especially true of an adult verte-
brate, but it is also true of one which is developing. The
chief difference is that, in the developing organism, no
- organ system can act upon another until it has itself begun
to function.
It is now proposed to explore the nature of the func-
tional interactions of those mammalian organ systems
which are important in development and such structural
elements, especially of the nervous system, as have signifi-
cance. The elements of the neuromuscular mechanism,
which appear to be the sole source of the initial activity of
the embryo, will be considered first. Not only are they of
supreme importance for an understanding of embryonic
activity, but there is perhaps more known about them
than concerning almost any other component of the
younger embryos.
The morphological development to functional capa-
city of the muscles, one of the two components of the
neuromuscular mechanism of the embryo, is fairly well
understood. The mesenchymal cells, from which the mus-
cles arise, develop into mononucleated spindle-shaped
myoblasts. As these elongate, their originally rather homo-
geneous cytoplasm exhibits longitudinal striations, the
myofibrillae, which are at first clear (Straus and Weddell,
1940, in the rat), shortly become beaded, and finally are
transformed into the definitive striated fibrils of skeletal
muscle. The myofibrillae tend to be very few in number at
first and to be located peripherally. Toward the center of
Structural and Functional Interrelationships in Prenatal Activity
[ 8 9 ]
the fiber the relatively large nuclei increase in number,
probably by their amitotic division (LeGros Clark, 1945).
Whether or not myofibrillae are an essential compon-
ent of mammalian skeletal muscle cells before they may
exhibit contractility is not yet determined. The work of
Goss (1940) indicates that cardiac muscle in the rat em-
bryo begins its contractions while in a completely undiffer-
entiated state. However, the fact remains that skeletal
muscle fibers, when they first become functional, possess
myofibrillae, although in very small numbers (Straus and
Weddell, 1940, and unpublished observations on the Pitts-
burgh material).
As noted in the first chapter (p. 6 ) , muscle activity
may be classified as myogenic, neurogenic, or reflexogenic.
It is certain that no spontaneous myogenic activity occurs
in mammalian skeletal muscle. Whether such muscle may
be mechanically or electrically excited to contraction with-
out the intervention of motor nerves is as yet an incom-
pletely settled problem. Straus and Weddell (1940) be-
lieve that all skeletal muscle in rat embryos, capable of be-
ing electrically excited to contraction, has a nerve supply
and that the responses secured are not "myogenic" but
neurogenic.
Neurogenic activity of skeletal muscle may be elicited
only shortly before true reflexogenic action begins. As the
skeletal muscle tissue in any vertebrate tends to differen-
tiate in a progressively cephalocaudal direction, true re-
flexes may be elicited in the more cephalic (actually cer-
vical) portions of an embryo before they can be secured
from its more caudal elements.
The neural component of the neuromuscular mechan-
ism develops within the central nervous system and the
[ 9 0 ]
spinal ganglia. The motor neurons within the spinal cord
precede the spinal ganglion cells in their differentiation. In
all forms studied, motor fibers reach the primitive mus-
cle-forming cell groups before differentiation in the latter
has occurred. This is, of course, the reason why the neuro-
genic type of activity precedes, though by only a few hours,
the reflexogenic.
Reflex activity requires, at a minimum, an afferent
(sensory or receptor) and an efferent (motor or effector)
neuron. Some proprioceptive reflex arcs are believed to be
bineuronal, but exteroceptive arcs are considered to be
multineuronal, having one or more association neurons
intercalated between the receptor and the effector cells
(Lloyd, 1943a, 1943b). It seems probable, though proof is
lacking, that it is the synaptic connections which are the
last elements of the arc to become functional in develop-
ment and, thus, establish the integrity of the reflex arc.
It seems doubtful that the spinal ganglion cells begin
to function as early as the cells of the semilunar (Gasser-
ian) ganglion, as Windle (1934a, 1934b) believes they do.
Angulo (1951a, 1951b) has shown that rat embryos of
264 hours fertilization age exhibit advanced fibrillation in
the trigeminal ganglion, but that no neurofibrillae are pres-
ent in the cervical ganglia before 288 hours. Responses to
exteroceptive stimulation in the rat do not appear before
378 hours, according to Angulo. Thus the semilunar gang-
lion cells are well in advance of the spinal ganglion cells in
differentiation in the rat. Much the same condition ap-
pears to exist in the human embryo.
Some authorities (Ranson, 1943; Clark, 1947) have
denied that mammalian embryos can exhibit a re-
sponse pattern basically similar to that of Amblystoma be-
[ 9 1 ]
cause of the absence of intramedullary ganglion cells in
the mammalian spinal cord. That intramedullary sensory
ganglion cells are actually present in the human embryonic
spinal cord has been amply shown both by Humphrey
(1944) and Youngstrom (1944), independently. Hum-
phrey demonstrated two generations of these primitive
intramedullary sensory neurons, the second of which is
more numerous at the beginning of activity than at any
other time. As they are predominantly present in the more
caudal portions of the cord, they can play no role in the
initial neck, trunk, and upper limb responses. However,
Humphrey has suggested (1952a) that they may well par-
ticipate, perhaps with the spinal ganglia, in the return to
posture of the pelvic region in slightly older embryos.
There appears to be no other evidence of the functioning
of the spinal ganglion cells in the earliest responses. In-
deed, all available indications point to the trigeminal nerveN
as mediating these earliest exteroceptive reflexes. '
The work of a considerable number of investigators
substantiates the view that external stimuli are first per-
ceived in the facial region. Some term it the snout, some
the perioral region, but all of these are merely brief ways
of denoting that region of the face supplied by the tri-
geminal nerve.
The trigeminal nerve terminations are probably the
earliest sensory receptors of the body (Windle, 1933;
Angulo, 1936). The problem has been to trace the reflex
pathway from V to the final common path in the spinal
cord. The difficulty here has been to determine, morpho-
logically, the caudal extent in the spinal tract of V of each
of its divisions through the medulla and into the spinal
cord. The fifth cranial nerve rarely stains or infiltrates with
[ 9 2 ]
silver in a differential manner, contrasting it with other
neighboring structures. Furthermore, animal operative ex-
perimentation and clinical postoperative observations have
led to contradictory results.
It has been generally accepted, as a result of functional
losses from accidental or operative injuries to the spinal
tract of V, that mandibular division (V3) fibers leave the
tract to enter its spinal nucleus in its most cephalic part,
that maxillary division (V2) fibers leave next and that
ophthalmic division (Vi) fibers alone extend to the caudal
limits of the tract. Among those whose work led to this
concept are Bregmann (1892) and Wallenberg (1896),
who carried out experimental tract sections on rabbits, and
a considerable group of clinicians who largely studied acci-
dental "natural experiments" injuring the tract, including
Schlesinger (1895), von Solder (1899), Kutner and Kra-
mer (1907), Spiller (1908), van Valkenburg ( 1 91 1 ) ,
Winkler (1921), Stopford (1924), Sjoqvist (1938), and
Smyth (1939).
The exact point of termination of the fibers of V2 and
V3 varies somewhat as given by different individuals. How-
ever, the majority opinion is that the fibers of these two di-
visions leave the descending root of V above the beginning
of the spinal cord. Bregmann, van Valkenburg, Winkler,
and Sjoqvist have stood almost alone in believing that V2
and V3 may extend to or into the first cervical segment
( Ci ) of the spinal cord. Actually, van Valkenburg presents
material that can be interpreted to indicate that some V2
and V3 fibers may pass as far caudally in the descending
tract as C3 in man, though he does not so state in his des-
criptions.
There is some evidence suggesting that the representa-
[ 9 3 ]
tion of each division may vary according to the animal
form. Barcroft and Barron (1939a) doubt that the fibers
of V3 in the sheep contribute any component to the des-
cending tract of V on the basis of their finding that this
division does not participate in the early trunk reflexes.
They propose a pathway for the earliest reflexes of the
sheep, which are otherwise quite similar to those described
in the Pittsburgh studies for man, as follows: fibers of V2
through the semilunar ganglion into the descending spinal
tract of V, thence into the reticular formation to establish
contact with the cells of origin of the reticulo-spinal tract
which, in turn, synapses with the lower motor neurons at
varying levels of the spinal cord. This is not only an en-
tirely possible reflex pathway from face to cervical cord,
but one for which there is considerable morphological evi-
dence. Furthermore, it does not require any further cau-
dalward extension of V
2
in the spinal tract than that pro-
vided by the work of previous investigators.
Recently, the morphological studies of Humphrey
(1951,1952a) have provided an alternate and more prob-
able pathway for the earliest reflexes elicited by perioral
stimulation. In a detailed survey of the Pittsburgh collec-
tion of sectioned tested human embryos, three of them
showed a differential silver impregnation which makes the
spinal tract of V clearly identifiable, marking it off from the
other longitudinal tracts of the dorsolateral quadrants of
the medulla and spinal cord.
The fibers of the three divisions of V have a well-known
arrangement (Bregmann, 1892, and others) in the spinal
tract of V, the fibers of V
3
being most dorsal, those of V2
in the middle and of V
x
ventral. Thus, Humphrey found
it possible to determine the general position of the fibers
[94]
of each of the divisions of V, even though they were not
differentially colored by the silver. Furthermore, there is
a persistence in adult man of the lateral notch in the spinal
root of V which marks the phylogenetically ancient fusion
of the originally separate ophthalmic nerve with the maxil-
lomandibular fibers. Thus, although it is only possible to
distinguish the general fiber area belonging to V2 from that
of V3, the maxillomandibular complex can readily be dis-
tinguished from the fiber mass of V
t
.
Because the arrangement of the ophthalmic and maxil-
lomandibular fiber groups in the spinal tract of V is ob-
viously the same in the human embryo as has been demon-
strated for the adult, it was possible for Humphrey to plot
the area occupied by Vi and by V2-V3 in the spinal tract of
V throughout its length and to establish their caudal limits
in each of the three embryos which showed differential sil-
ver impregnation of the trigeminal fibers. These three
were: 22 mm. CR (8 weeks), 25 mm. CR (8 weeks)
and 26.5 mm. CR (8.5 weeks), the age period during
which responses first appear on stimulation of the maxillo-
mandibular distribution of V to the face.
Humphrey's findings
7
may be summarized as follows:
1 ) in the 22 mm. CR embryo, Vi fibers end in the upper
part to the middle of C2, but the fibers of the V2-V3 com-
plex may be traced to C3; 2) in the 2 5 mm. CR embryo, V
x
fibers again end in mid-C2, the V2-V3 fibers extending to
lower C2 or upper C3; 3) in the 26.5 mm. CR embryo, Vi
fibers still end in mid-C2, but the V2-V3 fibers extend to or
into C4. The caudal extent of each fiber group differs
slightly on the right and left sides of the embryo (fig. 5)
and it must be borne in mind that, even though V
1
remains
near mid-C2 throughout this age period, its fibers have
FIG. 5 . Diagram of the caudal extent of the ophthalmic and maxil-
lomandibular fibers in the spinal tract of the trigeminal nerve in
young human embryos. (Courtesy of Dr. Tryphena Humphrey.)
grown in length to keep pace with the increase in length
of the embryo as a whole.
Thus, at the age when the maxillomandibular fibers of
the trigeminal nerve are functioning as the sensory limb
of the reflex arc for contralateral flexion on stimulation of
the perioral area, these divisions are already well into the
spinal cord. Indeed, they already extend farther caudally
than has ever before been demonstrated satisfactorily.
However, are there any connections between V2 and V3
and the ventral horn motor cells? Humphrey finds a wealth
of commissural fibers from the region of the spinal nucleus
[ 9 5 ]
CI
C2
C3
C4
ME D U L L A
REGION 22MM-8WKS. 25MM-8+WKS. 26.5MM-8.5WKS>
^ OP HT HA L . MAX. - MAN. OPHTHAL . I MAX. - MAN. OPHTHAL. I MAX. - MAN.
CN.S. R. L. R. L. R. L. R. L R . I R I
[96]
of V to the opposite ventral horn columns in Ci , C2, and
C3, but with the largest number in Ci and C2. Further-
more, some collaterals run to the nucleus of XI . Ipsilateral
fibers are also present, passing from the spinal nucleus of
V to the ventral motor columns and nucleus of XI on the
same side. It is interesting to note that Fitzgerald and
Windle (1937) state that ipsilateral responses predomi-
nate in the cervical level, although in their figure 10, of a
5-week human embryo, there are numerous contralateral
commissural fibers from the region of the spinal nucleus of
V, but only a small number of fibers passing to the same
side. This figure of Fitzgerald and Windle demonstrates
that the commissural connections of V long antedate
their being called into active function.
The reflex pathway presented by Humphrey for the
earliest contralateral reflexes in human embryos in re-
sponse to stimulation in the perioral region is as follows
(fig. 6 ) : 1) via the maxillomandibular divisions of V,
through the semilunar ganglion and the spinal tract, to the
spinal nucleus of Vi n the cervical cord; 2) thence predom-
inantly by commissural fibers to the ventral motor cell
columns of the opposite side, but with occasional ipsi-
lateral discharges to these columns, and slightly later to the
nuclei of XI on both sides; 3) then by cervical lower
motor neurons to the neck and upper trunk muscles to
produce the chiefly contralateral flexion in this region and
by the neurons of XI on both sides to facilitate the bilateral
extension of the arms at the shoulders.
Inasmuch as the maxillomandibular fibers of V are
present in the cervical spinal cord and the commissural
connections exist at the time when contralateral cervical
flexion begins, movements during the early age period
[ 9 7 ]
probably are mediated over this path. During later stages,
however, it seems likely that the reticulospinal pathway
suggested by Barcroft and Barron reinforces the response
and increases its caudal extent in the cervical and thoracic
regions of the spinal cord.
It would thus appear that there is a very definite reflex
pathway which indicates that the activities observed in
early human embryos and described in the preceding chap-
ter (pp. 62-68) are true exteroceptive reflexes mediated
over a morphologically demonstrated pathway. These ex-
teroceptive reflexes are basic to both pre- and postnatal ac-
tivities.
It now becomes necessary to examine the type or types
of reflexes exhibited by the embryo in its initial responses
and the pathways mediating them. Windle (1934b, 1944)
and Windle and Fitzgerald (1937) believe that the initial
reflex arcs to become functional in cat and human embryos
lie exclusively in the cervical cord, reaching its peak of de-
velopment in the brachial plexus region. It will be recalled
(pp. 36,47) that Swenson (1929) and Windle (1936) and
his co-workers are of the opinion that behavior in the mam-
mal has its origin "in simple, non-integrated, spinal type
reflexes" (Windle, p. 33P). Windle (1944 and elsewhere)
has stated that the earliest limb reflexes elicited by him in -
various mammals by tapping the amnion or "flipping" a
limb are proprioceptive reflexes but that the head reflex in
response to amnion tapping is exteroceptive. If the limb
flip response is a reflex, it is unquestionably of the stretch
type and it might be maintained that the head reflex on
amnion tapping could, at least sometimes, be of that type
also. Any change in head-body relationships, even of an in-
finitesimal amount, may stretch the neck muscles and
FIG 6 Diagram of the probable reflex pathway for contralateral
neck and upper trunk reflexes in the early human embryo. (From
Humphrey, 1952a, by permission of the Wistar Institute of An-
atomy and Biology.)
cause a proprioceptive reflex of this kind. All stretch re-
, flexes are narrowly localized, and Windle's head responses
from amnion tapping are as localized as his limb reflexes
from the same kind of stimulation. Only those head re-
[ 9 8 ]
[99]
sponses secured by Windle and others in response to stim- s
ulation in the snout region can, with assurance, be termed
exteroceptive. However, in the sheep, at least, there may
be doubt that the limb flip response is a reflex. Barron
(1944) has shown that reflex arcs are immature, if not in-
complete, in this form when such motility is exhibited.
There has been much misunderstanding of Sherring-
ton's (1906) term "the simple reflex." Even Coghill
(1909) leaned somewhat toward the simple reflex as an
entity, at first, but soon abandoned it as contrary to his
findings. Because of the misunderstanding, it seems well
to quote Sherrington's exact words (pp. 7-8):
The simple reflex. There is the co-ordination which a reflex
action introduces when it makes an effector organ responsive to
excitement of a receptor, all other parts of the organism being sup-
posed indifferent to or indifferent for that reaction. In this grade of
co-ordination the reflex is taken apart, as if separable from all other
reflex actions. This is the simple reflex. A simple reflex is probably
a purely abstract conception, because all parts of the nervous sys-
tem are connected together and no part of it is probablv ever cap-
able of reaction without affecting and being; affected bv various
other parts, and it is a system certainly never at rest. But the simple
reflex is a convenient, if not a probable, fiction. Reflexes are of
various degrees of complexity, and it is helpful in analyzing com-
plex reflexes to separate from them reflex components which we
may consider apart and therefore treat as though they were simple
reflexes.
Sherrington's discussion of the simple reflex clearly in-
dicates its mythical character. Even in the adult, the most *
localized reflexes, of the stretch type, can oocurJ only
through the co-ordinated action of other nervous mechan-
isms, such as those which inhibit the response of antagon-
istic muscles. This is equally true of the embryo, despite
the less complicated state of the nervous system in com-
parison with that in the adult.
[ 1 00]
According to the Windle concept, proprioceptive re-
flexes are not only early, they are the first to appear. How-
ever, the Pittsburgh studies have shown that definitely ex-
teroceptive reflexes also appear in man at the age in which
Fitzgerald and Windle (1942) first found proprioceptive
reflexes. Thus it would appear that proprioceptive reflexes
d!o not precede those of exteroceptive nature inbeing mani-
fested, although they may appear almost simultaneously.
Such an early appearance of proprioceptive reflexes
may be indicative of their increasing importance in higher
vertebrates^ but there is no evidence that behavior patterns
are built up from proprioceptive reflexes. On the con-
trary, all available evidence indicates that overt behavior
is primarily the response of the organism as a whole to its
external environment. This response is based on exterocep-
tive sense, with proprioceptive or interoceptive reactions
as incidental modifying factors. Thus, even if the Swenson-
Windle concept of unitary reflexes were to be accepted, it
is very doubtful, especially if they are proprioceptive in
nature, that they ever form the basis for the elaboration of
behavior since they involve only a part of the organism.
Windle, Becker, Rhines, and Cowgill (1942) and
Windle (1944,1950b) have challenged the normal nature
of all contralateral reflexes by the statement that responses
are normally ipsilateral and become contralateral only as a
result of asphyxia. This statement does not coincide with
the findings of the Pittsburgh studies or with those of An-
gulo (1932b). Responses to stimulation in the perioral (or
snout) region are very definitely contralateral in the vast
majority of cases and are of this character from the begin-
ning. Ipsilateral responses may occur following perioral
stimulation, but they are relatively infrequent in both rat
[101]
and man. Stimulation outside the perioral region in early
reactive ages always produces ipsilateral responses. These
are definitely caused by mechanical stimulation of the mus-
cle tissue directly. There is no known mechanism for the
reversal of this type of response into a contralateral form.
This statement by Windle leads directly to a considera-
tion of the oxidation of tissues and related matters. After
about the third week of fertilization age and the fifth of
menstrual age, the embryo has a blood circulation, for the
heart begins its beat at approximately that time. Prior to
the initiation of the heart beat, some other mechanism
must care for the oxygen needs of the dividing egg and the
developing early embryo, as oxygen diffusion alone is in-
adequate. This mechanism appears to be found in the ac-
tivities of the so-called oxidation-reduction enzymes. These
enzymes are known to be widely present, particularly in
early developing tissues, but are also active in later life. In-
deed, although they appear to decrease in activity as de-
velopment proceeds, little is known of the time at which
they cease to have an effect, if such cessation ever occurs.
All animal experiments have revealed a quite universal ap-
plicability of the chemical principles governing oxidation-
reduction enzymes.
These enzymes facilitate three types of chemical reac-
tions within the organism: "preparation" or conversion
reactions, those concerned with dehydrogenation, and
those aiding decarboxylation. Conversion reactions enter
into the alteration of 2-carbon compounds, from which
hydrogen cannot be separated, into 6-carbon compounds
from which hydrogen may be lost. These reactions prepare
embryonic compounds, especially for dehydrogenation,
and alter the chemicals of food materials at almost all ages.
[102]
Dehydrogenation reactions assist in the liberation of hy-
drogen ions which are absorbed by the pyridine nucleo-
tides, but their end result is to effect oxidation and to pro-
vide water. Decarboxylation reactions tend to liberate
CO2. In their activities, oxidation-reduction enzymes play
an important role in glycolysis, which is an essential feature
of nervous system respiration. Too little is known of the ef-
fectiveness of these enzymes once a circulation has been
established. However, Ball (1952) has shown that they do
act for adult nervous and other tissues under anoxial con-
ditions.
Once a circulation is established, at about five weeks of
menstrual age in the human embryo, it becomes one of the
important elements in the internal environment of the em-
bryo as a whole, as well as of the neuromuscular mechan-
ism. The circulation is most important in any considera-
tion of the activity of mammalian, and especially human,
embryos. Here, more than in any other class of vertebrates,
there is not only the danger, but the demonstrable fact, of
greater or less interference with normal O2-CO2 inter-
changes in all of the tissues of the embryo. It is only be-
cause they are metabolically more active than other tis-
sues or organ systems that nerve and muscle are particularly
affected.
In the mammalian embryo and fetus, the O2-CO2 in-
terchange, which occurs in the lungs of the postnatal in-
dividual, must take place in the placenta. Here the inter-
change of these gases between the embryo's own blood and
the maternal blood takes place through the placental bar-
riers interposed between the two blood streams. The ma-
ternal blood, in turn, is "aerated" in the mother's lungs.
The placental barriers, just mentioned, consist of the
[103]
layers of tissue separating the fetal from the maternal
blood. The number of layers differs in different forms and
the placentae of all mammals may be classified into five
categories on the basis of the layers present (Grosser, 1927;
Mossman, 1937). In spite of the differing number of layers
of the placental barriersfrom one to six, three in man-
all placentae function on the same principle, although it
appears that the thinner the barrier, the more efficient is
the placenta (Barron, 1951).
As Barcroft (1947) points out, the oxygen demand of
the fetal cells removes this gas from the fetal blood. This
O2 utilization lowers the oxygen pressure of the fetal blood
below that of the maternal blood. In consequence, "there
is thus a pressure gradient of oxygen foetus-wards" in the
placenta (see also, Barron, 1946). Barcroft also calls at-
tention to the fact that embryonic and fetal blood are
quite different from the blood of the postnatal infant.
Wintrobe and Shumacker (1936), working with the fe-
tuses, prematures, and neonates of mammals, including
man, have demonstrated that the red blood cells in early
fetuses of all forms studied are unusually large in size, that
most are nucleated, and that the hemoglobin per cell is
high. At this time, the total number of red blood corpus-
cles and the total amount of hemoglobin are both rela-
tively low. As development proceeds, the size of the red
cells and the percentage of nucleated reds decrease, but
the total amount of hemoglobin and the total number of
red blood corpuscles increase. However, the hemoglobin
per cell drops, so that the rise in total hemoglobin is not
great. Barcroft (1947) calls attention to the general agree-
ment among those who have studied fetal blood that the
hemoglobin content, hence the oxygen capacity, of fetal
[104]
blood rises during development, particularly in early em-
bryos, younger than the smallest (51 mm. CR, about 1 1 Vi
weeks) studied by Wintrobe and Shumacker. The initial
rise in the amount of hemoglobin is rapid at first and then
gradually slows to the relatively slight increment observed
by Wintrobe and Shumacker. This means, of course, that
the oxygen capacity of the early embryo is very low, in the
sheep less than a third of that of the late fetus. In conse-
quence, the amount of oxygen which embryonic blood is
capable of carrying is low normally, at the time of begin-
ning movements. Hence, the oxygen demand of embryos
and young fetuses must be very small, far less than in older
fetuses. This, also, is in agreement with the findings of
many workers in this field, such as Windle (1950a), Bar-
ron (1950), and others.
The low oxygen capacity of the early embryo is evi-
dently without untoward effect upon the activity of the
primary reflex arcs. They have developed in a still lower
oxygen environment and the greater supply at the time of
their beginning function is entirely adequate. This is not
the case, however, with newer neural mechanisms which
become active later. It is particularly not the case in con-
nection with those centers which have their synapses at
higher levels of the central nervous system. There is a very
definite gradient of oxygen demand in the central nervous
system, which increases in the cephalic direction. Cannon
and Burket (1913) gathered information determining the
following survival times, under complete anoxial condi-
tions, for the adult central nervous system: cerebrum, 8
minutes; cerebellum, 13 minutes; medulla, up to 30 min-
utes; and spinal cord, up to 60 minutes. This has been
amply substantiated (Himwich, 1951, pp. 288-290). In ad-
[105]
dition all newer neural components which develop find a
decreasing increment in oxygen supply, as mentioned in
the last paragraph. Although growth is well taken care of in
this respect, function is, at first, an added load not too well
tolerated. As a result, there is a tendency for newer neural
mechanisms at higher levels to succumb relatively early to
the effects of anoxia and asphyxia. This causes a reversion
in the behavior pattern to an earlier type of activity in
young fetuses. The tendency to reversion has been found
to be true for the rat by Angulo and has been noted in the
Pittsburgh human fetal studies.
In primary reflexes, on the contrary, and particularly in
those of exteroceptive type, quite other conditions prevail.
Here the afferent fibers are of the smallest caliber. These
are resistant to asphyxia, both in the experience of the
Pittsburgh group and as indicated by Windle et al.
(Windle and Becker, 1940; Fitzgerald and Windle, 1942;
Windle, Becker, Rhines and Cowgill, 1942; Windle,
1950b) and by Houssay (1951). In view of the fact that
the spinal cord has the lowest oxygen requirement of any
region of the central nervous system, exteroceptive reflexes
which are dependent on the cord for their synaptic trans-
mission, as are those over the trigeminal nerve (fig. 6 ) ,
should be the last to succumb to anoxia.
However, the effect of anoxia and asphyxia on reflexes
is twofold. Not only are reflexes ultimately abolished by
these conditions, but initially they are also facilitated by
them. Sherrington (1910) was probably the first to note
that a certain amount of asphyxia, in an adult cat, may
facilitate the eliciting of reflexes. More recently, Brooks
and Eccles (1947) have demonstrated that it is the anoxial
component of asphyxia which produces hyperexcitability,
[106]
as the excess of CO2 has a purely depressant effect. This
depression is, however, usually overridden, for a time, by
the excitatory effect of the anoxia. That anoxia may facili-
tate reflexes in mammalian embryos has been noted by
Angulo (1930b), Barron (1941), and Windle, Becker,
Rhines, and Cowgill (1942). A more detailed analysis of
the effects of anoxia on reflexes and on the development of
early fetal activity is to be presented by Humphrey
(1952b), and the reader is referred to her discussion for
this and other aspects of the subject.
Windle et aJ. (1940, 1942) have reported an initial
period, following immediately upon opening of the uterus,
during which the cat embryo failed to respond to stimu-
lation until the "brilliantly red" blood "began to darken"
within a matter of seconds. Then responses to stimulation
could be elicited and spontaneous activity soon appeared,
particularly if the placenta of the cat fetus under examina-
tion was removed from the uterus. This initial period of
no-response in the cat has been interpreted by these inves-
tigators as indicating that, under normal conditions in the
intact animal, fetal movements do not occur and that early
movements are only elicitable under the influence of the
f acilitatory action of anoxia.
However, it is by no means certain that early mammal-
ian and human movements do not occur normally in utero.
As far as man is concerned, it is a matter of common know-
ledge among obstetricians that, although the "quickening"
is usually said to begin about the 16th to the 18th week of
menstrual age, some patients "feel life" in the uterus as
early as the 10th to the 14th week of a pregnancy for which
the date of confinement is known to have been correctly
estimated.
[107]
Actually, it is unimportant whether intra-uterine ac-
tivity does occur normally in utero in the intact organism
or not, as the nervous
(
system continues to develop in either
the normal or enforced absence of movements. Harrison
(1904) placed frog embryos in a chloretone solution when
swimming reflexes were first appearing and kept them an-
esthetized for 7 days. Seventeen minutes after being again
placed in clear tap water, the embryos were capable of pro-
ducing normal reactions for their age. Herrick and Coghill
(1 91 5), Coghill (1924a), and Matthews and Detwiler
(1926) used chloretone solution on Amblystoma embryos
with similar results. It is true that when embryos were kept
for very long periods in a solution of chloretone, some in-
terference with the metabolism of the organism resulted
in slight retardation of development of portions of the
nervous system, causing some atypical responses under
these conditions. This is beside the point, as chloretone is
a highly abnormal environment. These experiments clearly
show that normal development may occur in the absence
of motility and, indeed, Coghill (1924b) points out that
the neuromuscular mechanisms for new types of reflexes,
however complex, develop without functioning until
"shunted" into the active system in a fully perfected state.
Concerning the initial no-response state reported by
Windle and Becker (1940) and Windle, Becker, Rhines,
and Cowgill (1942) for the cat, it must be borne in mind
that infraprimate mammals may differ greatly from pri-
mates in the response of their embryos. Some years ago, a
series of observations were made in Pittsburgh on rat
fetuses placed under conditions identical with those ob-
taining for the Pittsburgh human embryonic and fetal ob-
servations. The rat embryos or fetuses were removed with
[108]
the placenta, placed in warmed Tyrode's solution, and
stimulated by hairs. The results indicate that the rat fetus
does not act as does the human fetus of comparable de-
velopmental stages. Immediately following delivery, the
rat fetus exhibits under these conditions a very brief
period, limited to less than a minute, during which only
strong stimulation, interpreted as direct mechanical stimu-
lation of the muscle tissue, can evoke any movement. Such
responses as are elicited are entirely local. This may cor-
respond to Windle's period of no-response. Following this
no-response period, there is another, extending over a few
(2 to 3) minutes, during which reflexes of the exterocep-
tive type and corresponding to Angulo's (1932b) re-
sponses in rats of comparable age, may be secured. Reflexes
then disappear, but spontaneous movements, which begin
as soon as the reflexes, continue for periods of as long as
ten minutes.
The human fetus acts quite differently. Spontaneous
movements are not observed in all embryos, but those
which exhibit them do so ordinarily only within the first
minute after beginning placental separation. Embryos
which are capable of reflexes and which exhibit spontane-
ous movements apparently lack a prolonged period of no-
response. They react to external stimulation as soon as it is
applied, between one and two minutes after beginning pla-
cental separation. Those embryos which do not exhibit
spontaneous activity do have a very definite no-response
period, lasting up to three minutes or more. This is followed
by a period of response to external stimulation. By actual
count, only 31 per cent of the embryos and fetuses, con-
sidered as unanesthetized and from j
l
/
2
through 14 weeks
of age, exhibited a no-response period. These were believed
[109]
to be the cases suffering somewhat more trauma during ex-
traction than the 69 per cent which moved spontaneously
and had no period during which reflexes could not be se-
cured.
Exteroceptive responses may be evoked initially by
hairs having 2 5 mg., or higher, pressure values. In a few in-
stances, responses have been secured with 10 mg. hairs.
After a few (2 to 3) minutes, a 50 mg. or stiffer, hair is
necessary. In another minute or two, a 100 mg. hair is re-
quired. In other words, there is a steady rise in threshold
until responses cease. In 7%- to 8 Vi -week fetuses, this oc-
curs within 5 to 8 minutes after beginning placental sepa-
ration. Older fetuses respond over longer periods, up to
12 to 14 minutes at 14 weeks.
Whether Windle and Becker (1940) and Windle,
Becker, Rhines, and Cowgill (1942) are correct in assign-
ing the no-response period, when it occurs, to normal non-
motility or whether it is a phenomenon akin to surgical
shock, either in whole or in part, is undetermined. Indeed,
it does not greatly matter. Even if the earliest reflexes in
the embryo should only appear when facilitated by anoxia,
the character of such responses is unaffected by the anoxia
(Kato, 1934; Brooks andEccles, 1947; Humphrey, 1925b);
hence the neuromuscular mechanism exhibits activity
which is normal in character even under these conditions.
Furthermore, anesthesia which does not obliterate a pri-
mary reflex entirely has little effect upon the nature of the
reflex. If one includes all responsive human embryos and
fetuses from 7
x
/z through 14 weeks, instead of eliminating
those under suspicion of anesthesia, it appears that 30 per
cent showed a no-response period and 70 per cent ex-
hibited none, but moved spontaneously and reacted as
[110]
soon as tested. The fact that these figures correspond with
those from the selected nonanesthetized group is an added
bit of evidence confirming the contention of Kato (1934)
and others that anesthesia, short of obliterating nerve con-
duction, has little effect upon the normal character of a re-
sponse. Anesthesia may, however, delay the response or de-
crease the amplitude of the reaction as the anesthetic
blocks transmission in some of the fibers within the nerve.
The Pittsburgh studies have thus demonstrated an un-
broken series of continuously developing responses in man
from jVz weeks to the time when breathing can be estab-
lished. These may be interpreted as indicating the normal
sequence of neuromuscular potentialities for activity in
human embryos and fetuses at the different ages. Except
for the appearance of tendon reflexes after the establish-
ment, even temporarily, of respiration, there are no quali-
tative differences in the responses of breathing and non-
breathing fetuses of the same age. All responses in the
series are normal steps in the development of behavior,
whether or not they are somewhat facilitated by anoxia, or
whether or not the earliest occur in utero.
* * * * * *
The circulatory, muscular, and nervous systems of the
mammalian embryo are the first to reach a morphological
level which permits their beginning to function. In addi-
tion, the presence of oxidative-reductive enzymes aids tis-
sue respiration. Regarding other organ systems, relatively
little is definitely known, but that little should be noted,
even though more or less in passing, if for no other reason
than to indicate that their function may have significance.
That any of these other organ systems affects early mam-
malian overt behavior, as such, appears very doubtful at
[ 1 1 1 ]
this time, but their functioning forms a part of the internal
environment of the organism, even though they may con-
tribute little to covert behavior in the first half or more of
gestation.
It has been generally assumed that, because the pla-
centa carries out for the embryo and fetus the functions of
lungs, urinary system, and the absorptive aspects of the
digestive system, there is no need for these organs to begin
function in the developing organism until shortly before or
at the beginning of the viable period. Actually, these or-
gans exhibit function of a sort as soon as they are capable
of it.
The kidneys of the fetus begin to function at an early
age. Guthmann and May (1930) proved the presence of
urea and uric acid in small quantities in the amniotic fluid
of a human fetus of about the tenth week. In the Pitts-
burgh studies, the bladder has sometimes been found to
contain fluid at older ages (e.g., 15 Yi, 20Y i and 2 5 weeks).
This fluid was proved to be a dilute urine in the 25-week
specimen. In general, activity of fetuses causes emptying
of the bladder, so that fluid is relatively seldom found at
autopsy. Windle (1940) gives an excellent review of the
literature on this subject. However, it appears hardly pos-
sible that kidney function can affect the fetus seriously,
since the placenta can care for fetuses which lack kidneys,
ureters, or bladder (Preyer, 1885, pp. 326 et seq.).
Alimentary canal function also occurs in mammalian
fetuses. There is considerable doubt as to when various di-
gestive juices are formed, but none regarding the facts of
swallowing, peristalsis, and defecation. Preyer (1885) re-
viewed earlier studies. Becker, Windle, Barth, and Schulz
(1940) have clearly demonstrated, by the injection of
[112]
thorotrast into the amniotic fluid, that the guinea-pig
fetus begins to swallow on about the 42nd day (gestation.
67-69 days), exhibits transport of the material through
the alimentary canal, slowly at first, but with increasing
speed as term is approached, and begins to defecate into
the amniotic fluid about the 60th day. Yanase (1907)
demonstrated peristalsis in mammalian and human
fetuses. Although the Pittsburgh studies have demon-
strated that the human fetus may swallow as early as i2
L
A
weeks, it was not until 1946 that Davis and Potter demon-
strated the presence of thorotrast, injected into the am-
niotic fluid, in the alimentary canal of human fetuses as
early as 75 mm. in length (13 weeks). Windle and Bishop
(1939) and Becker et al. (1940) are of the opinion that
fetal anoxia tends to stimulate earlier deglutition, and
Windle has criticized the results observed by Davis and
Potter on the basis of fetal asphyxia. As yet, there is no
evidence that the human fetus normally defecates in utero.
Here, again, it seems unlikely that alimentary function
seriously affects the overt activities of the fetus.
The question of normal mammalian intra-uterine res-
piration is still in dispute, despite the fact that discussion
of this phenomenon began long ago. Preyer (1885), who
added personal observations of his own, notes that Vesa-
lius observed respiratory movements in fetuses of dog and
pig, and that a not inconsiderable number of early eight-
eenth century observers of mammalian fetuses, including
those of man, have recorded their data on respiration. In
the early observations, speculation was rife, as Preyer
points out, regarding the cause or causes of early mam-
malian respiration, whether it was anoxia and excess of
CO2, external stimulation, or both working together.
[113]
Ahlfeld and his associates, over a period of years (1888
to 1905), observed certain rhythmic activities of the hu-
man fetus in utero and in 1905 summarized the results ob-
tained and the conclusions drawn. These fetal movements
had a rate of 38 to 80 per minute at different ages, and were
not synchronous with the material aortic beat or respira-
tion, but were distinct from the abdominal movements of
the mother. Ahlfeld therefore concluded that these fetal
activities were respiratory movements. Although criticism
of Ahlfeld's work was severe, others also reported some-
what similar results (e.g., Reifferscheid, 1 91 1 ) .
A considerable number of observers have reported
finding lanugo, bits of vernix caseosa, and other detritus
from the amniotic fluid in the respiratory passages of still-
born infants or those which had died shortly after birth.
Such findings are not clear-cut evidence of normal intra-
uterine respiration, however, although they aid in demon-
strating that the respiratory apparatus can function to
some extent at very early ages.
Barcroft and Barron (1936) found that mere handling
of the sheep fetuses could start rhythmic respiratory type
movements between 38 and 49 days (gestation 157 days).
Such movements could not be excited at this stage by as-
phyxial conditions of the blood. The rhythm continued
for relatively long periods of time, when once begun, but
disappeared after the 50th day unless the umbilical cord
were clamped. In early stages, at least, there was no aspira-
tion of amniotic fluid, as negative pressure was not estab-
lished in the lungs.
Snyder, with Rosenfeld (1936, 1937a, 1937b, 1937c)
inhibited parturition in the rabbit and found that fetuses
between 28 and 34 days (gestation 32 days) exhibited
[114]
spontaneous rhythmic respiratory activity. Such activity
could be suppressed by injections of phenobarbital sodium
into the fetus. They concluded that intra-uterine respi-
ration of amniotic fluid during the latter part of pregnancy
is a normal occurrence in the mammalian fetus, indeed,
an essential step in the development of the lungs, as the
hydrostatic effect of inspired amniotic fluid serves to mold
the alveoli. In this latter opinion they are in agreement
with Wislocld (1920) and Bremer (1935). Snyder and
Rosenfeld (1937a) state that anoxia depresses and ulti-
mately abolishes these rhythmic activities, but that CO2
is necessary to maintain them, although the amount re-
quired is not critical.
Windle and his associates (1938 to 1950) have carried
out a considerable number of observations and experi-
ments on fetal respiration. Steele and Windle (1939) were
earlier convinced that, in the cat, fetal breathing shortly
before term might be a normal intra-uterine procedure, but
Windle, Becker, Barth, and Schulz (1939) later deter-
mined in the guinea-pig fetus that fetal intra-uterine res-
piratory movements are executed only under asphyxial
conditions and are not accompanied by aspiration of am-
niotic fluid in all cases.
Opposed to this is the work of Davis and Potter
(1946), who observed inspiration of thorotrast-laden am-
niotic fluid into the lungs of human fetuses between 75
and 300 mm. CR, which were about to be removed from
the uterus for well-substantiated medical reasons. As al-
ready noted, this work has been adversely criticized by
Windle.
In view of the unreconciled evidence presented, more
work in this field may be expected. About all that has been
[115]
clearly proved is that the respiratory apparatus is capable
of some degree of function at a very early fetal age in man
and other mammals.
In spite of the enormous number of contributions in
the literature, present-day knowledge regarding the time in
fetal life at which each of the endocrine organs may be-
come functional is still confused and incomplete. The situ-
ation is further complicated by the fact that at least some
hormones of maternal origin unquestionably pass through
the placenta from mother to fetus, and no criterion exists
for the separation of these maternal hormones from those
elaborated by the fetus itself. Efforts have been, and are
being, directed toward gaining further information regard-
ing the time at which fetal endocrine organs may begin to
elaborate their own secretions, the fetal age at which each
hormone becomes available in sufficient quantity to affect
its target organ, and the stage in development at which the
target organ may be affected by hormones of either fetal
or maternal origin.
As yet, the methods used in securing such information
have proved inadequate in most cases for an exact solution
of the problems presented. Some of the means used are
morphological, some have employed implantation of fetal
organs in immature postnatal animals to test their potency,
and some have used feeding tests or microchemical me-
thods of analysis. Each has its pitfalls, but each provides
some information which may throw light in the dark cor-
ners.
As neither space nor the purpose of this inquiry per-
mits an exhaustive review of the information even now
available, only a few references to the subject, illustrating
our present knowledge in the field, will be given here. The
[116]
interested reader may refer to the excellent reviews of
Moore (1947,1950) for a detailed account and the refer-
ences to the literature.
The goal of the morphological studies has been to de-
termine by histological examination, often with special
stains or impregnations, when the cytological structure of
endocrine cells reaches a level of development resembling
that of adult cells known to produce secretion and thereby
to estimate the fetal age at which these cells might secrete.
Except insofar as such cytological examination meets the
requirements of appropriate microchemical tests, there
is no evidence available that the attainment of a given
level of differentiation can be accepted as corresponding
with the beginning of function. In muscle tissues, nervous
tissues, and some other organs, the attempt to estimate
functional level from cytological structure has proved de-
ceptive and unreliable. Nevertheless, in the case of some
endocrine organs, there has been a reasonably close agree-
ment between the results of histological estimations and
of tests for physiological activity. In others, there has been
wide discrepancy.
In the pituitary gland, morphological studies indicate
that the alpha cells of the pars distalis achieve the ap-
pearance of functional capacity at varying fetal ages in dif-
ferent forms. In man, this appearance is reached early, at
about 12 weeks, but no tests for physiological activity have
as yet been made. The same is true in marsupials, where
the histologically estimated age of initial functional pos-
sibility is about 11 or 12 days, one or two days after the
young reach the pouch. In both these forms, the cyto-
logical evidence indicates a very early possible beginning
of secretion. However, in the pig and guinea pig, morpho-
[117]
logical evidence points to a midfetal date for the possible
beginning of function and these animals are born in a
more advanced condition of development than are mar-
supials and man. In the case of the pig, implantation or
feeding tests of physiological activity of the fetal pituitary
indicate that gonadotropic hormone is not present until a
week or two after the midfetal period and thyrotropic hor-
mone is not effective until shortly before birth. It must, of
course, be borne in mind that the time of beginning secre-
tion and the time of production of secretion in sufficient
quantity to be effective on a postnatal target organ may
well be widely separated from one another.
In the developing thyroid, colloid has been demon-
strated in the Pittsburgh embryos at 10 weeks of men-
strual age, but again tests for physiological activity are
lacking. In the rat, thyroids from 17-day fetuses failed to
metamorphose amphibian larvae, but those from 18-day
fetuses did so, and radioactive iodine has been shown to be
stored in the thyroids of 19-day fetuses.
The androgenic, or fetal, cortex or X-zone of the hu-
man suprarenal is very large beginning with the third fetal
month, but begins involution at about normal term. Invo-
lution is not completed until toward the end of the neo-
natal period. However, no tests for physiological activity
of the human suprarenal androgenic zone have been suc-
cessful during either the fetal or neonatal period. It is in-
teresting to note that the X-zone disappears from the
suprarenal of prematurely born infants within about 30
days. In some cases the premature attains the postneonatal
state a month or more before it would have done so had
it remained in utero until normal term. The question na-
turally arises, therefore, whether the supply of maternal
[118]
hormone passed to the fetus through the placenta main-
tains the adrenal androgenic zone until birth. Reports on
adrenal medullary function in human and other mammal-
ian fetuses are conflicting, and work on the generalized
cortex has failed to produce evidence of secretion.
Studies of pancreatic islet function have established
that insulin is present in the pancreas of the 5-month-old
calf in somewhat greater proportional amounts than in the
adult organ. Aside from this one demonstrated fact, puzzl-
ing contradictions appear. What appeared to be conclu-
sive evidence that insulin secreted by dog fetuses might
pass into the maternal blood and improve the condition
of the depancreatized diabetic mother (Carlson and Dren-
nan, 1 91 1 ) received a rude shock from the work of Cuth-
bert et al. (1940), who substantiated the earlier work, but
also showed similar maintenance of the bitch during nurs-
ing of the pups.
Histological studies have also proved confusing because
of the differences found in various forms. Alpha cells are
said not to be present before birth in the rat, the beta (in-
sulin-secreting) cells existing alone. Yet Donaldson and
Humphrey (1949) have shown that argentophilic cells
(considered by Ferner, 1938, and Hultquist, 1946, to be
alpha cells) are present in the human pancreas during the
9th week of menstrual age. Aggregates, containing both
argentophilic and other cells, considered to be islet tissue
(Humphrey and Donaldson, 1949) are present at 10
weeks.
The interstitial cells of the human testis are more
prominent in the 14- or 15-week fetus than in the adult,
but the exact stage at which they may begin secretion is
still unknown. Study of the so-called freemartin has indi-
[119]
cated, however, that their function probably begins early.
Cattle twins almost invariably form connections between
the two fetal membranes, with a continuity between the
chorioallantoic blood vessels of each resulting. Where one
twin is male and the other female, the latter's internal re-
productive system exhibits abnormalities of development
which fail to occur if both twins are females. The freemar-
tin has female external genitalia, but the internal organs
become largely those of the male sex. As testicular develop-
ment in the male precedes the ovary in differentiation, it is
believed that male secondary sex hormones secreted by the
interstitial cells of the male twin effect the maintenance of
the male organs of the female twin and the degeneration
or atrophy of her female organs. A somewhat similar ef-
fect is sometimes seen in the pig, but has not received the
careful study given the freemartin by Lillie (1917) and
others. The results of the injection of androgens and estro-
gens into pregnant female animals has further confounded
confusion. Androgens might be expected to produce free-
martin-like changes in female offspring when so injected
and do, with certain differences. However, estrogens,
similarly injected, tended to produce results somewhat
similar to those of androgens. To top this, the feeding ex-
periments of Warkany showed that maternal low vitamin
A diets tended to produce results similar to those secured
by estrogens and androgens.
In this account, only the surface of the contribution of
endocrine organs to fetal internal environment has been
touched. That the endocrine organs profoundly modify
behavior in postnatal life is clear, but the beginnings of
their activity and hence of their effects upon prenatal be-
havior are still so uncertain that more adequate treatment
[120]
here of an intensely interesting field has seemed neither
necessary nor wise.
* * * * * *
Evidence so far gathered appears to indicate that the
behavior of vertebrate animals, including man, has its
genesis in the early exteroceptive responses exhibited dur-
ing embryonic, larval, or fetal life. It would seem probable
that there is a basic similarity in the succession of responses
which may be elicited from vertebrate organisms in re-
sponse to external stimulation. However, it is evident that
each class of vertebrates, perhaps each order, genus, and
species, exhibits characteristics in the development of be-
havior which belong to that subdivision of animals alone.
Nevertheless, each form of activity shown by any fetal or-
ganism is a step in normal development, hence a step in
preparation for postnatal behavioral capabilities. Further-
more, there is a tendency for voluntary acts, where and
when they appear, to develop in a sequence based upon
the earlier reflexogenic sequence of prenatal life. This is
particularly well demonstrated in the case of human be-
haviorf The frontiers of research in the development of fe-
tal activity are by no means closed. Final proof of the
exact nature of many steps in the development of prenatal
activity in many forms is yet to come, but it would appear
that a reasonably sound foundation has been laid by the
work of many investigators, upon which future studies in
the field may rest.
Notes
1. See C. Judson Herrick's George Ellett Coghill, 1949, a mine
of information regarding the man and his work.
2. In this connection, see Herrick, 1949, p. 238, note 42.
3 . R. G. Harrison's staging of Amblystoma embryos 011 the
basis of morphological characteristics has, unfortunately, never
been published as a whole, but photographs of Miss Lisbeth
Krause's beautiful drawings of each stage have been widely distrib-
uted to investigators. Harrison's stages differ from CoghilTs, as the
latter were based on physiological criteria.
4. Dr. Coghill has objected (1940, p. 44, note) to the accep-
tance of Swenson's thesis as a published source, inasmuch as it was
never approved in toto for publication, some changes being desir-
able. However, as it is the only present source of Swenson's earlier
views, it is here used with the qualification just stated.
5. For another, and excellent, review of vertebrate behavioral
development, the reader is referred to Carmichael (1946).
6. The term "case" is used to denote each pregnancy termi-
nated spontaneously or by operation, regardless of whether the
"case" furnished one or more fetuses.
7. The writer wishes to express to Dr. Tryphena Humphrey
his thanks for permission to use in the third Porter Lecture pre-
sented at Kansas City on March 13, 1951, lantern slides of her
diagrams of the caudal extent of the divisions of the trigeminal
nerve and of the connections found to form the reflex arc in the
earliest responses of human embryos to exteroceptive stimulation
in the perioral region. These diagrams were presented by Dr. Hum-
phrey (1951) at the session of the American Association of Anat-
omists in Detroit on March 22, and some will be published in her
article (1952a) in the Journal of Comparative Neurology. One,
hitherto unpublished, is here presented as figure 5.
[121]
Aclmowledgnieiits
The writer takes pleasure in acknowledging with thanks per-
mission to make quotations or to reproduce illustrations from pub-
lished books or journals owned by the following:
To the Wistar Institute of Anatomy and Biology. Philadel-
phia, for excerpts from the Journal of Cellular and Comparative
Physiology, the Journal of Experimental Zoology and the Journal
of Comparative Neurology, including figures 4 and 6, tables 1, 2,
and 4. and use of the Swenson-Coghill film of Amblystoma, and to
the authors, Drs. Angulo, Barron, Humphrey, Sawyer, and Windle;
to the University of Chicago Press, Chicago, for quotations from
George Ellett Coghill by C. Judson Herrick, 1 9 4 9 , and for ex-
cerpts from Physiological Zoology, and to the authors, Drs. Herrick
and Windle; to The Journal Press, Provineetown, Massachusetts,
for quotations from Genetic Psychology Monographs, including
table 3 , and to the authors, Drs. Carmichael and Coronios; to the
Cambridge University Press, Cambridge, England, for excerpts
from the Journal of Physiology and to the author, Dr. Windle; and
to the Yale University Press for a quotation from Integrative Ac-
tion of the Nervous System by C. S. Sherrington, 1 906, and to Sir
Charles Sherrington, the author.
It is also a pleasure to acknowledge my debt to my colleagues,
Drs. Tryphena Humphrey and John C. Donaldson, to Dr. Donald
H, Barron, and to my wife, Helen Ferris Hooker, for their kindness
in criticizing and correcting the manuscript at various stages of its
completion, to Dr. Donaldson for his photographic assistance with
the illustrative material, and to Dr. Humphrey for the use of figure
5 , hitherto unpublished.
This book owes much to the wise counsel of Mr. Clyde K.
Hyder, of the University of Kansas Press. His keen editorial eye
and his knowledge of book production have been of great help to
the author, who is deeply grateful.
[122]
Reference List
AHLFELD, F., 1 905, Die intrauterine Tatigkeit der Thorax- und
Zwerchfelmuskeln. Mon. f. Geb. u. Gynak., 21:143-163.
ANGULO ( Y GONZALEZ) , ARMANDO W. , 1 927, The motor nuclei in
the cervical cord of the albino rat at birth. }. Comp. Neur.,
43 : 1 1 5 - 1 4 2 .
, 1 929, Is myelinogeny an absolute index of behavioral ca-
pacity? J. Comp. Neur., 48: 459-464.
, 1 930a, Neurological interpretation of fetal behavior: The
progressive increase of muscular activity in albino-rat fe-
tuses. Anat. Rec, 45: 254.
, 1 930b, Endogenous stimulation of albino rat fetuses.
Proc. Soc. Exp. Biol, and Med., 27: 579.
, 1 932a, The prenatal growth of the albino rat. Anat. Rec,
5 2 : 1 1 7 - 1 3 8 .
, 1 93 2b, The prenatal development of behavior in the al-
bino rat. J. Comp. Neur., 55: 395-442.
, 1 93 3 , Development of somatic activity in the albino rat
fetuses. Proc. Soc. Exp. Biol, and Med., 3 1 : 1 1 1 - 1 1 2 .
, 1 93 5 , Further studies upon development of somatic activ-
ity in albino rat fetuses. Proc. Soc. Exp. Biol, and Med.,
3 2 : 621 - 622.
, 1 936, The neuro-mechanism of the trunk limb component
of the total behavior pattern. Anat. Rec, 64, Suppl. No. 3:
2-3.
, 1 93 9, Histogenesis of the monopolar neuroblast and the
ventral longitudinal path in the albino rat. J. Comp. Neur.,
7 1 : 325-360.
, 1 95 1 a , A comparison of the growth and differentiation of
the trigeminal ganglia with the ganglia of the cervical re-
gion in albino rat embryos. Anat. Rec, 1 09; 391 .
, 1 9 5 1 b, A comparison of the growth and differentiation of
the trigeminal ganglia with the cervical spinal ganglia in
albino rat embryos. }. Comp. Neur., 95: 53-71 .
AVERY, GEORGE T., 1 928, Responses of foetal guinea pigs prema-
turely delivered. Genet. Psychol. Monogr., 3 : 245- 331 .
BAER, KARL E. VON, 1 828, Ueber Entwicklungsgeschichte der
Thiere. 2 vols., Konigsberg. (Cited from Preyer, 1 885. )
BALFOUR, FRANCIS M., 1 874, A preliminary account of the devel-
opment of the elasmobranch fishes. Mem. Edit. Works of
F. M. Balfour, 1 : 6 0 - 1 1 2 , 1 8 8 5 .
, 1 878, A monograph on the development of the elasmo-
[123]
[124]
branch fishes. Mem. Edit. Works of F. M. Balfour, 1 : 203-
5 2 0, 1 885 .
BALL, ERIC G., 1 95 2, Oxidations and reductions in the brain tissue.
Biological Aspects of Mental Health and Disease, Chap.
6, pp. 74-82 (Milbank Memor. Fund Symposium, Nov. 1 3 -
1 6, 1 9 5 0) . Hoeber, New York.
BARCROFT, JOSEPH, 1 93 9, The intra-uterine development of respi-
ratory effort. Brit. Med. J. , Nov. 1 8 , 1 9 3 9 : 986-987.
, 1947 Researches in pre-natal life. Chas. C. Thomas,
Springfield, 111.
BARCROFT, JOSEPH, AND D. H. BARRON, 1 93 6, The genesis of res-
piratory movements in the foetus of the sheep. J . Physiol.,
88: 56-61.
, 1 937, Movements in midfoetal life in the sheep embryo.
J . Physiol., 91 : 329- 351 .
, 1 939a, Movement in the mammalian foetus. Ergebn. d.
Physiol., biol. Chem. u. exper. Pharmacol., 42: 1 07- 1 52.
, 1 939b, The development of behavior in foetal sheep.
J . Comp. Neur., 70: 477-502.
BARCROFT, J., D. H. BARRON AND W. F. WI NDLE, 1 936, Some ob-
servations on genesis of somatic movements in sheep em-
brvos. J . Physiol., 87: 73-78.
BARRON, DONALD H., 1 9 4 1 , The functional development of some
mammalian neuromuscular mechanisms. Biol. Rev., 16:
1 -3 3 .
, 1 944, The early development of the sensory and inter-
nuncial cells in the spinal cord of the sheep. J . Comp. Neur.,
81 : 1 93 - 2 2 5 .
, 1 046, The oxygen pressure gradient between the maternal
and fetal blood in pregnant sheep. Yale J . Biol, and Med.,
1 9: 23-27.
, 1 950, Genetic neurology and the behavior problem.
Genetic Neurology, ed. by P. Weiss: 223 - 23 1 , Univ. of
Chicago Press, Chicago.
, 1 9 5 1 , Placental morphology and fetal respiration. Science,
1 1 4: 477.
BECKER, R. F., W. F. WI NDLE, E. E. BARTH, AND M. D. SCHULZ,
1 940, Fetal swallowing, gastro-intestinal activity and de-
fecation in utero. Surg., Gyn. and Obst., 70: 603- 61 4.
BEGUEHN, 17157, Abhandlung von der kunstliche geoffnete Eier
beim Lampenfeur auszubriiten. Hamburg. Mag. 0. d.
gesam. Schrift a. d. Naturforsch. u. d. angen. Wissen.
uberh., 1 9; 1 1 8- 1 5 6. (Cited from Preyer, 1 885. )
BERSOT, H., 1 920- 1 , D6veloppment reactionnel et r^flexe plan-
Reference List
taire du h6b6 n6 avant terme a celui de deux ans. Schweiz.
Arch. f. Neurol, u. Psychiatr., 7: 21 2- 23 1 ; 8: 47-74.
BOELL, EDWARD J. , AND S. C. SHEN, 1 950, Development of cholin-
esterase in the central nervous system of Amblystoma punc-
tatum. J. Exp. Z00L, 1 1 3 : 583-600.
BOLAFFIO, M., AND G. ARTOM, 1 924, Richerche snlla fisiologia
del sistema nervosa del feto umano. Arch. d. Scienze biol.,
5: 457- 487.
BOYD, EDITH, 1 94 1 , Outline of physical growth and development.
Burgess, Minneapolis.
BREGMANN, E. , 1 892, Crber experimentelle aufsteigende Degener-
ation motorischer und sensibler Hirnnerven. Arb. a. d.
Inst. f. Anat. n. Physiol, d. C.n.s. a.cl. Wiener Univ., 1 :
73-97-
BREMER, JOHN L. , 1 93 5 , Postnatal development of alveoli in the
mammalian lung in relation to the problem of the alveolar
phagocyte. Carnegie Contrib. to Embryol., 2 5 : 83- 1 1 0.
BRIDGMAN, C. S., AND L. CARMICHAEL, 1 93 5 , An experimental
study of the onset of behavior in the fetal guinea-pig. J .
Genet. Psvchol., 47: 247-267.
BRINLEY, FLOYD J. , 1 9 5 1 , Studies on the effect of curare on spon-
taneous muscular activity in fish embryos. Physiol. Zool.,
24: 186-195.
BROOKS, C. M C C , AND J . E . ECCLES, 1 047, A study of the effects of
anesthesia and asphyxia on the mono-synaptic pathway
through the spinal cord. J . Neurophvsiol., 10: 349-360.
CAJ AL, SANTIAGO R., 1 906, Genesis delas fibras nerviosas del em-
bri6n v observaciones contrarias a la teoria catenaria. Tra-
bafos, Madrid, 4: 227-2Q4.
CANNON, WALTER B., AND I. R. BURKET, IQI 3 , The endurance of
anemia by nerve cells in the myenteric plexus. Am. J. Phy-
siol., 3 2 : 347-3 ^7-
CARLSON, A. J. , AND F. M. DRENNAN, i on, The control of pan-
creatic diabetes in pregnancy bv the passage of the internal
secretions of the pancreas of the fetus to the blood of the
mother. Am. J . Physiol., 28: 3 9 1 - 3 9 5 .
CARMICHAEL, LEONARD, 1Q34, An experimental study in the pre-
natal guinea-pig of the origin and development of reflexes
and patterns of behavior in relation to the stimulation of
specific receptor areas during the period of active fetal life.
Genet. Psychol. Monogr., 16: 337- 491 .
, 1 946, The onset and early development of behavior. Chap.
2 , Manual of Child Devel., ed. by L. Carmichael: 43-1 66.
Wiley, New York.
[125]
[126]
CLARK, ELEANOR L., AND E. R. CLARK, 1 91 4, On the early pulsa-
tions of the posterior lymph hearts in chick embryos:
Their relation to the body movements. J. Exp. Zool., 1 7 :
373-394.
CLARK, SAM L,, 1 947, Ranson's The anatomy of the nervous sys-
tem. 8th Ed. W. B. Saunders, Philadelphia.
COGHILL, GEORGE E., 1 902, The cranial nerves of Amblystoma
tigrinum. J. Comp. Neur., 1 2 : 205-289.
, 1 909, The reaction to tactile stimuli and the development
of the swimming movement in embryos of Diemyctylus
torosus Eschscholtz. J. Comp. Neur., 1 9: 83- 1 05.
, 1 91 4- 1 93 6, Correlated anatomical and physiological
studies of the growth of the nervous system of Amphibia.
J. Comp. Neur., I, 24: 1 61 - 2 3 3 ( 1 9 1 4 ) ; II, 26: 247-340
( 1 9 1 6 ) ; III, 3 7: 37-69 ( i 9 M
a
) ; IV, 3 7: 7 1 - 1 2 0 ( 1 92 4b) ;
V, 40: 47- 94 ( 1 92 6a ) ; VI, 4 1 : 9 5 - 1 5 2 ( 1 9 2 6 b) ; VII, 42:
1 - 1 6 ( 1 92 6c ) ; VIII, 45 : 227-247 ( 1 9 2 8 ) ; IX, 5 1 : 3 1 1 - 3 7 5
( 1 93 0a ) ; X, 5 3 : 1 47- 1 68 ( 1 9 3 1 ) . XI, 57: 327-358
( 1 93 3 a ) ; XII, 64:135-167 ( 1 9 3 6 ) .
, 1 929, Anatomy and the problem of behavior. Cambridge
Univ. Press, Cambridge, England.
, 1 930b, Anatomical growth of the nervous system in rela-
tion to behavior. Proc. Conf. on Adolesc., Publ. as: Physical
and mental adolescent growth: 1 06- 1 1 0,
1 930c, The structural basis of the integration of behavior.
Proc. Nat. Acad. Sci., 1 6: 637-643.
, 1 933b, Somatic myogenic action in embryos of Fundulus
heteroclitus. Proc. Soc. Exp. Biol, and Med., 32: 62-64.
, 1 940, Early embryonic somatic movements in birds and
in mammals other than man. Monogr. Soc. Res. Child De-
vel., 5: No. 2.
CORONIOS, J. D., 1 03 3 , Development of behavior in the fetal cat.
Genet. Psychol. Monogr., 14: 283-386.
CUTHBERT, F. G., A. C. IVY, B. L. ISAACS, AND J. GRAY, 1 940, The
relation of pregnancy and lactation to extirpation diabetes
in the dog. Am. J. Physiol., 1 1 5 : 480-496.
DAVIS, M. EDWARD, AND EDITH L. POTTER, 1 946, Intrauterine res-
piration of the human fetus. J. A. M. A., 1 3 1 : 1 1 9 4 - 1 2 0 1 .
DONALDSON, JOHN O, AND TRYPHENA HUMPHREY, 1 949, Cells
with argentophilic granules in the embryonic human pan-
creas. Anat. Rec, 1 03 : 532- 533.
EMMERT, A. G. F., AND HOCHSTETTER, 1 8 1 1 , Die Entwicklung
der Eidechsen in ihren Eiern. Arch. f. d. Physiol., 1 0: 86,
95 , 1 00- 1 04, 376. (Cited from Preyer, 1 885 . )
Reference List
[127]
ERBKAM, 1837, Lebhafte Bewegung eines viermonatlichen Fotus.
Neue Zeitschr. f. Geburtsk., 5: 324-326.
FERNER, HELMUT, 1938, Ober die Entwicklung der Langerhans-
schen Inseln nach der Geburt und die Bedeutung der versil-
berbaren Zellen im Pankreas des Menschen. Zeitsch. f.
mikr.-anat. Forschung, 44: 451-488.
FITZGERALD, JAMES E., AND W. F. WI NDLE, 1942, Some observa-
tions on early human fetal activity. J . Comp. Neur., 76:
159-167.
GESELL, ARNOLD, WITH C. S. AMATRUDA, 1945, The embryology
of behavior. Harper, New York.
Goss, C. M., 1940, First contractions of the heart without cyto-
logical differentiation. Anat. Rec, 76: 19-27.
GROSSER, OTTO, 1927, Frujientwicklung, Eihautbildung, und
Placentation. Bergmann, Munich.
GUTHMANN, H., AND W. MAY, 1930, Gibt es intrauterine Nieren-
sekretion? Arch. f. Gynak., 141:450-459.
HALVERSON, H. M., 1937, Studies of the grasping responses of
early infancy. J . Genet. Psychol., 51: 371-449.
, 1943, The development of prehension in infants. Child
Devel. and Behavior, ed. by Barker, Kounin, and Wright,
Chap. 4: 49-65. McGraw-Hill, New York.
HARREVELD, A. VAN, 1944, Survival of reflex contractions and in-
hibition during cord asphyxia. Am. J . Physiol., 141: 97-101.
HARRISON, Ross G., 1904, An experimental study of the relation
of the nervous system to the developing musculature in the
embryo of the frog. Am. J . Anat, 3: 197-220.
, 1910, The outgrowth of the nerve fiber as a mode of proto-
plasmic movement. J. Exp. Zool., 9: 787-848.
HARVEY, WI LLI AM, 1651, Exercitationes de generatione animal-
ium. Pulleyn, London. (See Meyer, 1936.)
HELD, HANS, 1909, Die Entwicklung des Nervengewebes bei den
Wirbeltieren. Barth, Leipzig.
HENSEN* V., 1903, Die Entwicklungsmechanik der Nervenbahnen
im Embryo der Saugetiere. Kiel.
HERRICK, C. JUDSON, 1949, George Ellett Coghill, Naturalist and
Philosopher. Univ. of Chicago Press, Chicago, 111.
HERRICK, C. JUDSON, AND G. E. COGHILL, 1915, The development
of reflex mechanisms in Amblystoma. J . Comp. Neur., 25:
65-85.
HEWER, EVELYN E., 1938, Textbook of histology for medical stu-
dents. Mosby, St. Louis.
HIMWICH, HAROLD E.> 1951, Brain metabolism and cerebral dis-
orders. Williams and Wilkins, Baltimore.
[128]
His, WI LHELM, 1 887, Die Entwicklung der ersten Nervenbahnen
beim menschlichen Embryo. Arch. f. Anat. u. Physiol,
Anat. Abt, Jahrg. 1 887, 368-378.
HOME, EVERARD, 1 8 2 2 , On tie changes the egg undergoes during
incubation. Philos. Trans. Roy. Soc. Lond., 2 : 339-356.
HOOKER, DAVENPORT, 1 936, Early fetal activity in mammals. Yale
J. Biol, and Med., 8: 579-602.
, 1 93 8, The origin of the grasping movement in man. Proc.
Am. Philos. Soc, 79: 597-606.
, 1 93 9, Fetal behavior. Res. Publ. Assn. Res. Nerv. and
Ment. Dis., 1 9: 237- 243.
, 1 942 , Fetal reflexes and instinctual processes. Psychosom.
Med., 4: 1 99- 205.
, 1 943 , Reflex activities in the human fetus. Child Behav.
and Devel., ed. by Barker, Kounin and Wright, Chap. 2:
1 7- 28. McGraw-Hill, New York.
, 1 944, The origin of overt behavior. Univ. of Mich. Press,
Ann Arbor.
HOUSSAY, BERNARDO A., 1 9 5 1 , Human physiology. McGraw-Hill
Book Co., Inc., New York.
HULTQUIST, GOSTA T., 1 946, On the occurrence of s. c. silver cells
in islet tumors and in the pancreas in cases of islet tumors.
Gastroenterologia, 7 1 : 1 93- 209.
HUMPHREY, TRYPHENA, 1 944, Primitive neurons in the embryonic
human central nervous system. J. Comp. Neur., 81 : 1-45.
, 1 9 5 1 , The caudal extent of the descending root of the
trigeminal nerve during the period of early human fetal
activity (8 to 8.5 weeks of menstrual age). Anat. Rec, 1 09:
306-307.
, 1 952a, The spinal tract of the trigeminal nerve in human
embryos between 7V2 and SY2 weeks of menstrual age and
in relation to fetal behavior. J. Comp. Neur., 96: 1-67.
, 1 952b, The relation of oxygen deprivation to fetal reflex
arcs and the development of fetal behavior. }. Psychol., 34
(in press).
HUMPHREY, TRYPHENA, AND J. C. DONALDSON, 1 949, Argento-
phile cells in the early embryonic human pancreas. Anat.
Rec, 1 03 : 572.
KATO, GENICHI, 1 93 4, The microphysiology of nerve. Mazuren,
Tokyo, pp. 1 3 9.
KERR, J. GRAHAM, 1 9 1 0 , The development of the peripheral nerves
of vertebrates. Proc Roy. Phys. Soc Edinburgh, 1 8: 1 1 - 20.
KRABBE, KNUD, 1 9 1 2 , Les reflexes chezle foetus. Rev. Neurol., 24:
434-435-
Reference List
Kuo, ZING Y. 1 93 2- 1 93 8, Ontogeny of embryonic behavior in
birds. I, J. Exp. Z00L, 61 : 395-430 ( 1 93 2 a ) ; II, J. Exp.
Z00L, 62: 453-487 ( 1 9 3 2 b ) ; III, J. Comp. Psychol, 1 3 :
245- 271 ( 1 9 3 2 c ) ; IV, J . Comp. Psychol, 1 4: 1 09- 1 22
( i 9
3
2 d) ; V, Psychol. Rev., 39: 499-515 ( i 93
2 e
) ;
V I
> J-
Comp. Psychol, 16: 379-384 ( 1 9 3 3 ) ;
x
(
w t m T
-
c
-
SHEN) , J. Comp. Psychol, 21: 87-93 l
1
9 3
6
) >
X I
(
w i t n
T.
C SHEN) , J. Comp. Psychol, 24: 49-58 ( 1 93 7 ) ; XII, Am.
J. Psychol, 5 1 : 361 -379 ( 1 9 3 8 ) .
* 1 93 9, Studies on the physiology of the embryonic ner-
vous system. IV. Development of acetylcholine in the
chick embryo. J. NeurophysioL, 2; 488-493.
KUTNER, R., AND F. KRAMER, 1 907, Sensibilitatsstorungen bei
acuten und chronischen Bulbarerkrankungen. Arch. f.
Psychiatr., 42 : 1 002- 1 006.
LANE, H. H., 1 9 1 7 , The correlation between structure and func-
tion in the development of the special senses of the white
rat. Univ. Okla. Bull, No. 1 40 (Univ. Studies No. 8 ) : 1-
88.
LANGREDER, WI LHELM, 1 949, Welche Fotalreflexe haben eine
intrauterine Aufgabe. Deutsch. Med. Wochenschr., 74:
661 -667.
L E GROS CLARK, W. E., 1 945, The tissues of the body. 2nd Edit
Clarendon Press, Oxford.
LIESCH, E., 1 946, La motilita riflessa durante lo sviluppo fetale
neir uomo. Boll. Soc. Ital Biol. Sper., 22: 831 - 833.
LI LLI E, FRANK R., 1 9 1 7 , The freemartin; a study of the action of
sex hormones in the foetal life of cattle. J. Exp. Zool, 2 3 :
371 - 452.
LLOYD, DAVID P. C, 1 943a, Reflex action in relation to pattern and
peripheral source of afferent stimulation. J. Neurophysiol,
6 : 1 1 1 - 1 1 9 .
, 1 943b, Conduction and synaptic transmission of reflex
response to stretch in spinal cats. J. Neurophysiol, 6: 31 7-
326.
MATTHEWS, S. A. , AND S. R. DETWILER, 1 926, The reactions of
Amblystoma embryos following prolonged treatment with
chloretone. J. Exp. Zool, 4 5 : 279-292.
MEYER, ARTHUR W. , 1 936, An analysis of the De Generatione
Animalium of William Harvey. Stanford Univ. Press.
MINKOWSKI, M. 1 920a, Reflexes et mouvements de la t&te, du
tronc et des extrmits du foetus humain, pendant la pre-
miere moitie de la grossesse. C. R. d. 1. Soc. Biol, 83: 1 202-
1 204.
[ 1 29 ]
[130]
, 1 920b, Movimientos y reflejos del feto humano durante la
primera mitad del emberazo. Trabajos, Madrid, 1 8: 269-
2
7 3 -
, 1 920c, Uber Bewegungen und Reflexe des menschlichen
Foetus wahrend der ersten Halfte seiner Entwicklung.
Schweiz. Arch. f. Neurol, u. Psychiatr., 7: 1 48- 1 5 1 .
, 1 9 2 1 , Sur les mouvements, les reflexes et les reactions mus-
culaires du foetus humain de 2 a 5 mois et leurs relations
avec le syst&me nerveux foetal. Revue Neurol., 37: 1 1 0 5 -
1 1 1 8 , 1 2 3 5 - 1 2 5 0 .
, 1 922, Uber frtihzeitige Bewegungen, Reflexe und musku-
lare Reaktionen beim menschlichen Foetus und ihre Be-
ziehungen zum foetalen Nerven und Muskelsystem.
Schweiz. med. Wochenschr., 3 : 721 - 724, 751 - 755.
-, 1 923 , Zur Entwicklungsgeschichte, Localisation und
Klinik des Fussohlenreflexes. Schweiz. Arch. f. Neurol, u.
Psychiatr., 1 3 : 475- 51 4.
, 1 925, Zum gegenwartigen Stand der Lehre von den Re-
flexen. Neurol, u. Psychiatr. Abhandl., Schweizer Archiv.
f. Neurol, u. Psychiatr., 1 : 1 - 61 .
, 1 92 4 - 5 , Zum gegenwartigen Stand der Lehre von der Re-
flexen. (in entwicklungsgeschichtlicher und anatomish-
physiologischer Beziehung) Schweiz. Arch. f. Neurol, u.
Psychiatr., 1 5 : 239-259; 16:133-152, 266-284.
, 1 926, Sur les modalitis et la localisation du r6flexe plan-
taire au cours de son 6volution du foetus a Tadulte. Congres
d. med. ali6n. et neurol. de France et d. pays d. lang. franc,
30; 301 -308.
, 1 927, t)ber du elektrische Erregbarkeit der foetalen Mus-
kulatur. Schweiz. Arch. f. Neurol, u. Psychiat., 2 2 : 64-72.
, 1 928, Neurobiologische Studien am menschlichen Foetus.
Handbuch d. biol. Arbeitsmethoden. E. Abderhalden, Abt.
V, Teil 5 B, Heft 5, Ser. Nr. 25 3 : 5 1 1 - 61 8.
-, 1 936, Cervical and labyrinthine reflexes in the human
fetus. (Original in Russian.) Symp. on problems of ner-
vous physiology and behavior, in honor of Prof. J. S. Beri-
tofL USSR Acad. Sci., Georgian Br.: 249-258.
, 1 946, Sur le d^veloppement et la localisation des fonctions
nerveuses, surtout des mouvements et des reflexes, chez le
foetus et le nouveau-n6. Atti d. Congr. Italo-Svizzero d.
Neuropsicol. Infantile: 35-57.
, 1 946, Sur T Evolution anatomo-physiologique des fonc-
tions c6r6brales chez le nourisson et le petit enfant. Atti d.
Congr. Italo-Svizz. d, Neuropsicol. Infantile: 59-80.
Reference List
[131]
- , 1 947, Sur le debut d'une evolution biopsychique chez le
foetus. C. R. d. Congr. d. Med. AH6n. et Neurol., 45th ses-
sion: 1 - 1 0.
MOORE, CARL R., 1 947, Embryonic sex hormones and sexual dif-
ferentiation. Chas. C. Thomas, Springfield, 111.
, 1 950, The role of the fetal endocrine glands in develop-
ment. J. Clin. Endocrin., 1 0: 942-985.
MOSSMAN, HARLAND W. , 1 937, Comparative morphogenesis of
the fetal membranes and accessory uterine structures. Car-
negie Contrib. to Embryol., 26: 129-246.
NEEDHAM, JOSEPH, 1 93 1 . Chemical Embryology. Cambridge
Univ. Press, 3 vols.
ORR, DOUGLAS W. , AND W. F. WI NDLE, 1 934, The development
of behavior in chick embryos: the appearance of somatic
movements. J . Comp. Neur., 60: 271 - 285.
PANKRATZ, D. S., 1 93 1 , A preliminary report on the fetal move-
ments in the rabbit. Anat. Rec, 48: Suppl., 58-59.
PATON, STEWART, 1 905, The reactions of the vertebrate embryo
to stimulation and the associated changes in the nervous
system. Mitt. a. d. zool. Stat, zu Neapel, 1 8: 535-581 .
, 1 9 1 1 , The reactions of the vertebrate embryo and asso-
ciated changes in the nervous system. Second paper. J.
Comp. Neur., 21: 345-373.
PFLUGER, E. F. W. , 1 877, Die Lebenszahigkeit des menschlichen
Foetus. Arch. f. d. ges. Physiol., 1 4: 628-629.
PREYER, WI LHELM, 1 822, Die Seele des Kindes, Beobachten
liber die geistige Entwicklung des Menschen in den ersten
Lebensjahren. Fernau, Leipzig.
, 1 885, Specielle Physiologie des Embryo. Grieben, Leipzig.
RANSON, S. WALTER, 1 943, The anatomy of the nervous system.
7th ed. W . B. Saunders, Philadelphia.
RAWITZ, BERNHARD, 1 879, Lebenszahigkeit des Embryo. Arch f.
Physiol., Suppl. Bd: 69-71.
REIFFERSCHEID, K., 1 9 1 1 , Ueber intrauterine Atembewegungen
des Foetus. Arch. f. d. ges Physiol., 1 40: 1 - 1 6.
REMAK, R., 1 854, Die Zusammenziehung des Amnions. Arch. f.
Anat., Physiol., u. wiss. Med.: 369-373. (Cited from Preyer,
1 885. )
ROSENFELD, MORRIS, AND F. F. SNYDER, 1 936, Foetal respiration in
the rabbit. Proc Soc Exp. Biol, and Med., 3 3 : 576-578.
SAWYER, CHARLES H., 1 934a, Cholinesterase and the behavior
problem in Amblystoma. I. The relationship between the
development of the enzyme and early motility. II. The ef-
fects of inhibiting cholinesterase. J . Exp. Zool., 92: 1-29.
, 1 943 b, Cholinesterase and the behavior problem in
Amblystoma. III. The distribution of cholinesterase in
nerve and muscle throughout development. IV. Cholines-
terase in nerveless muscle. J. Exp. Zool., 9 4 : 1 - 3 1 .
, 1 944, Nature of the early somatic movements in Fundu-
lus heteroclitus. J. Cell, and Comp. Physiol., 24: 71 - 84.
SCHLESINGER, H., 1 895, Die Syringomy61ie. Deutlicke, Leipzig.
SHERRINGTON, CHARLES S., 1 906, The integrative action of the
nervous system. Yale Univ. Press, New Haven, Conn.
, 1 9 1 0 , Notes on the scratch reflex of the cat. Q. J. Exp.
Physiol., 3 : 21 3- 220.
SJOQVIST, O., 1 938, Studies on pain conduction in the trigeminal
nerve. Acta Psychiatr. et Neurol., Suppl. No. 1 7 : 1 - 1 3 9.
SMITH, KARL U., AND ROBERT S. DANIEL, 1 946, Observations of
behavioral development in the loggerhead turtle (Caretta
caretta). Science, 1 04: 1 5 4- 1 5 5 .
SMYTH, G. E., 1 939, The systemization and central connections
of the spinal tract and nucleus of the trigeminal nerve. A
clinical and pathological study. Brain, 62: 41 -87.
SNYDER, FRANKLIN F. , AND M. ROSENFELD, 1 93 7a, Direct observa-
tion of intrauterine respiratory movements of the fetus and
the role of carbon dioxide and oxygen in their regulation.
Am. J . Physiol., 1 1 9: 1 53- 1 66.
, 1 93 7b, Breathing of amniotic fluid as a normal function of
fetal respiration. Proc. Soc. Exp. Biol, and Med., 3 6: 45-46.
, 1 937c, Intrauterine respiratory movements of the human
fetus. J. A. M. A., 1 08: 1 946- 1 948.
SOLDER, F. VON, 1 899, Der segementale Begrenzungstypus bei
Hautanaesthesien am Kopfe, insbesondere in Falle von
Syringomyelic. Jahrb. f. Psychiatr. u. Neurol., 1 8: 458-
478.
SPILLER, W . G., 1 908, The symptom-complex of occlusion of the
posterior inferior cerebellar artery: two cases with necropsy.
J. Nerv. and Ment. Dis., 3 5 : 365-387.
STEELE, A. G., AND W . F. WI NDLE, 1 93 9, Some correlations be-
tween respiratory movements and blood gasses in cat foe-
tuses. J. Physiol., 94: 531 - 538.
, 1 9 39, The oxygen and carbon dioxide content of the blood
of normal and pregnant decerebrate cats. J. Physiol., 94:
525-530.
STOPFORD, JOHN S. B., 1 924, The function of the spinal nucleus of
the trigeminal nerve. J. Anat., 59: 1 20- 1 28.
STRASSMAN, P., 1 903 , Das Leben vor der Geburt. Samml. Klin.
Vortr., no. 353 (Gynak. No. 1 3 2 ) : 947-968.
[132]
Reference List
STRAUS, WI LLI AM L., JR., AND G. WEDDELL, 1 940, Nature of the
first visible contractions of the forelimb musculature in rat
fetuses. J . Neurophysiol., 3 : 358-369.
STREETER, GEORGE L. , 1 920, Weight, sitting height, head size,
foot length and menstrual age of the human embryo. Car-
negie Contrib. to Embryol., 1 1 : 1 43- 1 70.
SWENSON, ENGELBREICT A., 1 926, The development of movement
of the albino rat before birth. Thesis, Univ. Kansas (un-
published).
, 1 928, The simple movements of the trunk of the albino
rat. Anat. Rec, 38: 3 1 .
, 1 929, The active simple movements of the albino rat
fetus: the order of their appearance, their qualities, and
their significance. Anat. Rec, 42: 40.
TI LNEY, FREDERICK, 1 93 3 , Behavior in its relation to the develop-
ment of the brain. II. Bull. Neur. Inst. N. Y., 3: 252- 358.
TI LNEY, FREDERICK, AND L. S. KUBIE, 1 9 3 1 , Behavior in its rela-
tion to the development of the brain. I. Bull. Neur. Inst.
N. Y , 1 : 229- 31 3.
TRACY, HENRY C, 1 925, Relation of C 0
2
to spontaneous move-
ments in the larvae of Opsanus tau. Biol. Bull., 48: 408-
4 1 1 .
, 1 926, The development of motility and behavior reactions
in the toadfish (Opsanus tau). J . Comp. Neur., 40: 253-
3 69.
TUGE, HIDEOMI, 1 9 3 1 , Early behavior of the embryos of the tur-
tle, Terrapene Carolina ( L . ) . Proc Soc. Exp. Biol, and
Med., 29: 52-53.
, 1 934, Early behavior of the embryos of carrier-pigeons.
Proc Soc Exp. Biol, and Med., 31 : 462-463.
, 1 93 7, The development of behavior in avian embryos. J .
Comp. Neur., 66: 1 57- 1 79.
VALENCIENNES, 1 841 , Observations faites pendant rincubation
d'une femelle du Python bivittatus. Annal. d. Sci. Natur.,
16: 65-72. (Cited from Preyer, 1 885. )
VALKENBURG, C. T. VAN, 1 9 1 1 , Zur Kenntnis der Radix spinalis
nervi trigemini. Monatschr. f. Psychiatr. u. Neurol., 29:
407-437.
VULPIAN, A., 1 85 7, La physiologic de Tamnios et d' allantoide chez
les oiseaux. M6m. 1. a 1. Soc. de Biol., Ser. 2, 4: 269-278.
(Cited from Preyer, 1 885. )
WALLENBERG, ADOLF, 1 896, Die secundare Bahn des sensiblen
Trigeminus. Anat. Anz., 1 2 : 95- 1 1 0.
WANG, GING HSI, AND T. W . LU, 1 940, Spontaneous activity of
spinal tadpoles of frog and toad. Science, 92: 1 48.
[133]
[134]
, 1 94 1 , Development of swimming and righting reflexes in
frog (Rana guentheri): Effects thereon of transection of
central nervous system before hatching. J. Neurophysiol., 4:
1 37- 1 46.
WEI SS, PAUL, 1 941 a Autonomous versus reflexogenous activity of
the central nervous system. Proc. Am. Philos. Soc, 84: 5 3 -
64.
, 1 94 1 b, Does sensory control play a constructive role in the
development of motor co-ordination? Schweiz. med.
Wochenschr., 71 s t year: 591 - 595.
, 1 941 c , Self differentiation of the basic patterns of co-
ordination. Psychol. Monogr., No. 88, 1 7 : 1-96.
WHI TE, GERTRUDE M., 1 9 1 5 , The behavior of brook trout em-
bryos from the time of hatching to the absorption of the
yolk sac. J. Anim. Behav., 5: 44-60.
WINDLE, WI LLI AM F. , 1 930, The earliest fetal movements in the
cat correlated with the neurofibrillar development of the
spinal cord. Anat. Rec, 45 : 249.
, 1 9 3 1 , The neurofibrillar structure of the spinal cord cor-
related with the appearance of early somatic movements. J.
Comp. Neur., 55: 7 1 - 1 1 3 .
1 93 3 , Neurofibrillar development in the central nervous
system of cat embryos between 8 and 1 2 mm. long. J.
Comp. Neur., 58: 643-723.
, 1 934a, Correlation between the development of spinal
reflexes and reflex arcs in albino-rat embryos. Anat. Rec,
58: Suppl. Am. Assn. Anat, 42.
, 1 934b, Correlation between the development of local
reflexes and reflex arcs in the spinal cord of rat embryos.
J. Comp. Neur., 59: 487-505.
, 1Q36, The genesis of somatic behaviour in mammalian
embryos. J. Physiol., 87: 31 P- 33P.
, 1 937, On the nature of the first forelimb movements of
mammalian embryos. Proc Soc. Exp. Biol, and Med., 36:
640-642.
, 1 940, Physiology of the fetus. Saunders, Philadelphia.
, 1Q44, Genesis of somatic motor function in mammalian
embryos: a synthesizing article. Physiol. Zool., 1 7: 247-260.
:
, 1 950a, Asphyxia neonatorum. Charles C. Thomas, Spring-
field, 111.
, i9^ob, Reflexes of mammalian embryos and fetuses. Gen-
etic Neurology, ed. by Paul Weiss, Univ. of Chicago Press:
214-222.
WI NDLE, W . F., AND R. E. BAXTER, 1 93 6, Development of reflex
. mechanisms in the spinal cord of albino rat embryos. Cor-
Reference List
relations between structure and function, and comparisons
with the cat and the chick. J. Comp. Neur., 63: 189-209.
WI NDLE, W . F. , AND R. F. BECKER, 1940, Relation of anoxemia
to early activity in the fetal nervous system. Arch. Neur.
and Psych., 43: 90- 1 01 .
WI NDLE, W . F. , R. F. BECKER, E. E. BARTH, AND M. D. SCHULZ,
1 93 9, Aspiration of amniotic fluid by the fetus. An experi-
mental roentgenological study in the guinea pig. Surg.,
Gyn. and Obst., 69: 705- 71 2.
WI NDLE, W . F., R. F. BECKER, RUTH RHINES, AND ELIZABETH J.
COWGILL, 1 942, Effects of anoxia on trigeminal reflexes in
cat fetuses. Physiol. Z00L, 1 5 : 375-382.
WI NDLE, W . F. , AND C. L. BISHOP, 1 939, Prenatal intestinal move-
ments in anoxia. Proc. Soc. Exp. Biol, and Med., 40: 2-4.
WI NDLE, W . F. , C. A. DRAGSTEDT, D. E. MURRAY AND R. R.
GREENE, 1 9 38, A note on the respiration-like movements of
the human fetus. Surg., Gynec. and Obst., 66: 987-988.
WI NDLE, W . F. , AND M. W . FISH, 1 023, The development of the
vestibular righting reflex in the cat. J . Comp. Neur., 54:
85-96.
WI NDLE, W . F., AND J . E. FITZGERALD, 1 937, Development of the
spinal reflex mechanism in human embryos. J . Comp.
Neur., 67: 493-509.
WI NDLE, W . F., AND A. M. GRIFFIN, icm, Observations on em-
bryonic and fetal movements of the cat. J . Comp. Neur.,
52: 14Q-188.
WI NDLE, W . F., W . L. MINEAR, M. F. AUSTIN, AND D. W. ORR,
1 93 5 , The origin and early development of somatic be-
havior in the albino rat. Physiol. Zool., 8: 1 56-1 8^.
WI NDLE, W . F., M. MONNIER AND A. G. STEELE, 1 938, Fetal res-
piratory movements in the cat. Physiol. Zool, 1 1 : 425-433.
WI NDLE, W . F., J . E. O' DONNELL, AND F. E. GLASSHAGLE.JLQ^,
The early development of spontaneous and reflex behavior
in cat embryos and fetuses. Physiol. Zool., 6: 521 -541 .
WI NDLE, W . F., AND D. W. ORR, 1 934, The development of be-
havior in chick embryos: spinal cord structure correlated
with early somatic motility. T- Comp. Neur., 60: 287-307.
WI NDLE, W . F., D. W . ORR, AND W . L. MINEAR, 1 934, The ori-
gin and development of reflexes in the cat during the third
fetal week. Physiol. Zool., 7: 600-617.
WI NDLE, W , F., AND A. G. STEELE, 1 938, Reciprocal oxygen
changes on both sides of placenta during uterine contrac-
tion and relaxation. Proc. Soc. Exper. Biol, and Med., 39:
246-248.
[135]
[136]
WINKLER, CORNELIS, 1 9 2 1 , Le systeme du nerf trijumeau. Opera
Omnia, 7: 1 - 1 00.
WINTERSTEIN, H., 1 9 1 4 , Ueberleben eines menschlichen Fotus
bei kunstliche Durchspiilung. Deutsch. Physiol. Gesell.,
Zentralbl. . Physiol., 28: 728.
WINTREBERT, P., 1 920, La contraction rythme aneurale des myo-
tomes chez lez embryons de s61aciens. I. Observation de
Scylliorhinus canicula L. Gill. Arch, de Zool. Exper. et
Gen., 60: 221 - 459.
WINTROBE, M. M. , AND PI. B. SHUMACJKER, JR., 1 936, Erythro-
cyte studies in the mammalian fetus and newborn. Am. J .
Anat, 58: 31 3- 328.
WISLOCKI, GEORGE B., 1 920, Experimental studies on fetal ab-
sorption. Carnegie Contrib. to Embryol., 1 1 : 45-60.
WOYCIECHOWSKI, B., 1 928, Rucky zarodka ludzkiego 42 mm.
Polsk. Gazeta Lekarska, 7: 409- 41 1 .
YANASE, J. , 1 907, Beitrage zur Physiologie der peristaltischen Be-
wegungen des embryonalen Darmes. Arch. f. d. ges. Phy-
siol, 1 1 7 : 345-383; 1 1 9 : 451 - 464.
YOUNGSTROM, KARL A., 1 93 8, Studies on the developing behavior
of Anura. J . Comp. Neur., 68: 351 - 379.
, 1 94 1 , Acetylcholine esterase concentration during the
development of the human fetus. J . Neurophysiol, 4: 473-
477-
, 1 944, Intramedullary sensory type ganglion cells in the
spinal cord of human embryos. J . Comp. Neur., 81: 47-53.
ZUNTZ, NATHAN, 1 877, Ueber die Respiration des Saugetierfoetus.
Arch. f. d. ges. Physiol, 1 4: 605-627.
Index of Names
(Italicized page numbers refer to items in the Reference List.)
Ahlfeld, F., 113, 133
Amatruda, C. S., 127
Angulo (y Gonzalez), A. W., 12, 34,
35, 36-39, 41, 84, 90, 91, 100, 105,
xo6, 108, 123
Artom, G., 55, 125
Austin, M. F., 45-46, 135;
Avery, G. T., 34, 123
Baer, K. E. von, 27,123
Balfour, F. M., 7, 123-124
Ball, E. G., 102, 124
Barcroft, J., 46, 48-51, 54, 67, 74,
7^ 93 97 103*104 113 124
Barron, D. H., 11, 12, 48-51, 67, 74,
78, 84, 93, 97, 103, 104, 106, 113,
124
Barth, E. E., 111-112, 114, 124, 135
Baxter, R. E., 134
Becker, R. F., 100, 105, 106, 107,
109, 111-112, 114, 124, 135
Beguelin, 27, 124
Bersot, H., 54, 124-12$
Bishop, C. L., 112, 135
Boell, E. J., 24, 125
Bolaffio, M., 55, 125
Boyd, E., 62, 125
Bregmann, E., 92, 93, 125
Bremer, J. L., 114, 125
Bridgman, C. S., 42-43, 12$
Brinley, F. J., 12, 125
Brooks, C. McC., 105, 109, 125
Bitrket, I. R., 104, 125
Cajal, S. R., 9, 12$
Cannon, W. B 104, 125
Carlson, A. J.
t
118, 125
Carmichael, L., 38, 39-43, 121, 125
Clarlc, E. L.
f
28, 126
Clark, E. R., 28, 126
Clark, S. L,, 90, 126
Coghill, G. E., 5, 9, 11-12, 14-21, 22,
23, 24, 25, 29, 34 3* 3
6
> 41 4">>
51-52, 53, 71, 84, 99, 107, 121,
126", 127
Coronios, T. D., 38-40, 126
Cowgill, E. J., 100, lot;, 106, 107,
109, 135
Cuthbert, F. C, 118, 126
Daniel, R. S., 25-26, 132
Davis, M. E., 54, 112, 114, 126
Detwiler, S. R., 107, 129
Dragstedt, C. A., 80, 135
Drennan, F. M., 118, 125
Donaldson, J. C, 118, 126, 128
Eccles, J. E., 105, 109, 125
Emmert, A. G. F., 25, 126
Erbkam, 54, 127
Ferner, H., 118,127
Fish, M. W., 66, 135
Fitzgerald, J. E., 55-57, 58, 64, 79,
96, 97, 100, 105, 127,
Gesell, A., 77, 127
Glasshagle, F. E., 43, 45, ^5
Goss, C. M., 89,127
Gray, J., 126
Greene, R. R., 80, 135
Griffin, A. M., 43-44, 135
Grosser, O., 103, 127
Guthmann, H., m'
127
Halverson, H. M., 84, 127
Harreveld, A. van, 127
Harrison, R. G., 9, 107, 121, 127
Harvey, W., 27
12
7> 129
Held, H., 9
12
7
Hensen, V., 9,127
Herrick, C. J., 5 17, 20, 25, 107,
121, 127
Hewer, E. E., 81, 127
Himwich, H. E., 104, 127
His, W., 9, 128
Hochstetter, 25, 126
Hogg, I. D., 57
Home, E., 27, 128
Hooker, D., 84, 128
Houssav, B. A., 105,
12
%
Hultquist, G. T., 118,
1 2
8
Humphrey, T., 57> 91, 9V97, 08 1
109, 118, 121, 126, 128
06,
Isaacs, B. L., 1 25
Ivy, A. C, 126
[137]
Index of Names
Kato, G., 1 09- 1 1 0, 128
Kerr, J . G. , 9, 128
Krabbe, K., 54,, 128
Kramer, F., 92, 129
Krause, L., 1 2 1
Kubie, L, S., 34, 1 33
Kuo, Z. Y. , 28-30, 32, 129
Kutner, R., 92, 129
Lane, H. H., 33, 34, 1 29
Langreder, W. , 54, 129
LeGros Clark, W. E. , 89, 129
Liesch, E. , 54, 129
Lillie, F. R., 1 1 9, 1 2 9
Lloyd, D. P. C., 90, 129
Lu, T. W. , 23, 133-134
Matthews, S. A. , 1 07, 129
May, M. , 1 1 1 , 1 2 7
Meyer, A. W. , 127, 129
Minear, W. L., 45-46, 1 35
Minkowski, M. , 54-55, 67, 129-131
Monnier, M. , 1 35
Moore, C. R., 1 1 6, 131
Mossman, H. W. , 1 03, 1 31
Murray, D. E. , 80, 1 35
Needharn, J . , 1 31
O'Donnell, J . E. , 43, 45, 1 35
Orr, D. W. , 31-33, 45-46, 1 31 , 1 35
Pankratz, D. S., 37-38, 41 , 1 31
Paton, S., 7 , 0, 131
Pfluger, E. P. W. , 54, 1 31
Potter, E. L., 54, 1 1 2 , 1 1 4, 126"
Preyer, W. , 5, 14, 2 1 - 2 2 , 2 ; , 27 - 2 8 ,
33, 34, 54, 1 1 1 , 1 1 2 , 1 31
Ranson, S. W. , 90, 131
Rawitz, B. , 5 4 . 132
Reifferscheid, K., 1 1 3 , 1 31
Remak, R., 27, 1 31
Rhines, R., 100, 1 05, 106, 107, 109,
-*35
Rosenfeld, M., 1 1 3-1 1 4 13*, *32
Sawver, C. H. , o. 1 2- 1 3, 2 4 , 1 3 1 - 1 3 2
Schlesinger, H., 0 2 , 1 3 2
Schulz. M. D., 1 1 1 - 1 1 2 , 1 1 4 . 1 2 4 .
Shen, S. C., 24, 1 25
Shen, T. C., 129
Sherrington, C. S., 99, 1 05, 132
Shumacker, H. B. , 103-104, 136
Sjoqvist, O., 92, 132
Smith, K. U., 25-26, 132
Smyth, G. E., 92, 132
Snyder, F. F., 1 1 3 - 1 1 4, ty>W
Solder, F. von., 92, 132
Spiller, W. G., 92, 132
Steele, A. G., 1 1 4, 1 32, 135
Stopford, J . S. B., 92, 132
Strassmann, P., 54, 132
Straus, W. L., Jr., 88, 89, 133
Streeter, G. L., 62, 1 33
Swenson, E. A. , 34, 35-36, 48, 52,
97, 1 2 1 , 133
Tilney, F., 34, 1 33
Tracy, H. C. , 9-1 1 , 1 2, 34, 35, 133
Tuge, H., 25, 28, 30-31, 32, 133
Valenciennes, 25, 1 33
Valkenburg, C. T. van, 92, 133
Vesalius, A. , 1 1 2
Vulpian, A. , 27-28, 1 33
Wallenberg, A , 92, 1 33
Wang, G. H. , 23, 133-134
Warkany, Josef, 11Q
Weddell, G., 88, 89, 133
Weiss, P . , 20, 22, 134
White, G. M. , 1 0, 134
Windle, W. F . . 28, 31 -33, 43-48, 5
2
"
53> 5^57* 5
8
>
6
4> 66, 79, 80, 90,
Q l , 96, 97, Q8, 9 0 , 1 00-1 01 , 1 0 4 ,
1 05, 106, I O 7 , 108, I O 9 , 1 1 1 - 1 1 2 ,
1 1 4, 1 2 4 , 127, 3 3 1 , 1 32, 134-135
Winkler, C., Q2, 1 36
Winterstein, H., $4, 136"
Wintrebert, P . , 7 - Q , 1 0, 1 1 , 1 3 6 "
Wintrobe. M. M. , 1 0 3 - 1 0 4 , 136
Wislocki, G. B., 1 1 4 1 3 6 "
Woyciechowski, B. , 54, 136"
Yanase, T.. = ? 4 , 1 1 2 , 136*
Youngstrom, K. A. , 22-23, 24, 84,
Zuntz, N. , i?4, 136
[138]
Subject Index
(Italicized page references are to figures or tables on the
pages cited.)
A c e t y l c h o l i n e , 30
A c r i s , 22, 84
A c t i v i t y , f e t a l , g e n e r a l , 3-4,16*, 87-88,
1 20
i n t r a - u t e r i n e , 106-107
m u s c u l a r , a n e u r a l , 7
m y o g e n i c , 5-13, 22, 27, 29, 30,
33, 47, 63, 89
" m y o g e n i c / ' 6, 1 4, 36-37, 89
m y o t o m i c , 7-8, 1 0- 1 1 , 14-17, 19,
n e u r o g e n i c , 6-13, 20, 22, 29, 31 ,
33> 37
8
9> 9
r e f l e x o g e n i c , s e e R e f l e x e s , e x t e r o -
c e p t i v e
s p o n t a n e o u s , 6-14, 20-23, 25, 26,
29, 3 3
1
* 3
2
> 33 4"4
6
> 4
8
>
49, 66, 76, 89, 106, 108-109
A l b i n o r a t , s e e R a t
A l i m e n t a r y s y s t e m , 1 1 1 - 1 1 2
A l l a n t o i c c o n t r a c t i o n s , s e e u n d e r
C h i c k
A m b l y s t o m a , 5, 1 2, 14-24, 34, 48, 84,
90, 1 07
A m n i o n , c o n t r a c t i o n s , s e e u n d e r
C h i c k
t a p p i n g , 43, 46, 47, 56, 64, 97-98
A m p h i b i a , 14-25, 42, 53, 1 1 7
A n e s t h e s i a , e f f e c t s , 48, 57-59, 107,
1 09- 1 1 0
e m b r y o n i c , 1 07, 1 09-1 1 0
m a t e r n a l , 57-59
A n o x i a , g e n e r a l , 47-48, 72, 1 1 2- 1 1 4
e f f e c t o n r e f l e x e s , 34, 47, 53, 56,
57, 65, 102-106, IO9, 1 1 2
f a c i l i t a t i o n b y , 1 0- 1 1 , 105-106, 109,
1 1 0
A n u r a , 22-25, 84
A s p h y x i a , e f f e c t s o f , 47, 49, 53, 69,
100, 105-106, 1 1 2, . 1 1 3
s u p p r e s s i o n b y , 1 1 , 71 , 105-106,
1 1 4
A t r o p i n , e f f e c t s o f , 58-59
A v e r t i n , e f f e c t s o f , 5 8
B a r b i t u r a t e s , e f f e c t s o f , 58, 1 1 4
B e h a v i o r , g e n e r a l , 3-4, 100, 1 1 0, 120
c o v e r t , 3, 6 , 87-88, 1 1 0- 1 20
o v e r t , 3-4, 6 , 53, 87, 100, 1 1 0- 1 1 1
B i r d s , 22, 27-33
B u f o , 22-24, 84
C a l f , 1 1 8
C a r e t r a carettst, 25-26
C a r b o n d i o x i d e , g e n e r a l , 1 0 - n , 74,
1 02, 105-106, 1 1 2 , 1 1 4
e n d o g e n o u s s t i m u l a t i o n b y , s e e u n -
d e r S t i m u l a t i o n
f a c i l i t a t i o n b y , s e e u n d e r A n o x i a
s u p p r e s s i o n b y , 1 1 , 74, 105-106
C a t , 34, 38, 40, 43-46, 66, 97, 1 05,
106, 107, 1 1 4
C a t s h a r k , 7-10
C a t t l e , 1 1 8- 1 1 9
C e n t r a l n e r v o u s s y s t e m , g e n e r a l , 9,
24, 104, 1 05
t r a n s e c t i o n o f , 23
C h i c k , g e n e r a l , 27-30, 31-33
a l l a n t o i c c o n t r a c t i o n s , 27, 33
a m n i o t i c c o n t r a c t i o n s , 22, 27-29,
3> 33
y o l k s a c c o n t r a c t i o n s , 27
C h l o r e t o n e , e f f e c t s o f , 107
C h o l i n e s t e r a s e ( C h E . ) , 1 2- 1 3, 23-24,
30
C o n t r a c t i o n s , s e e A c t i v i t y , m u s c u l a r
C o o l i n g , e f f e c t s o f , 34
C o r a m i n e , 58
C u n n e r , 9, 10
C u r a r e , 1 2, 30-31
D e g l u t i t i o n , s e e S w a l l o w i n g
D e m e r o l , 58-59
D i a p h r a g m a t i c m o v e m e n t s , 44, 49,
74, 81
D o g , 1 1 2 , 1 1 8
E n d o c r i n e s y s t e m , 1 1 5- 1 20
E n d o g e n o u s s t i m u l a t i o n , s e e u n d e r
S t i m u l a t i o n
E n v i r o n m e n t , e x t e r n a l , 3, 87-88, 100
i n t e r n a l , 3, 87-120
E n z y m e s , o x i d a t i o n - r e d u c t i o n , 1 01 -
1 02, 1 1 0
E p h e d r i n e , 58, 59
E s e r i n e , 1 3
[139]
Subject Index
E s t e r a s e , 1 3
E s t h e s i o m e t e r , 1 4, s e e a l s o S t i m u l a -
t o r s
E t h e r , 58
E y e m o v e m e n t s , 29, 49, 71 , 76, 83
F a c i a l n e r v e , s e e N e r v e , c r a n i a l , V I I
F i s h e s , 7-13, 20, 21 , 22, 34
F l o o r - p l a t e c e l l s , 1 7, 18
F r e e m a r t i n , 1 1 8- 1 1 9
F r o g , s e e R a n a
Fundulus, 9-13
G a s p i n g , 74, 80
G u i n e a p i g , 34, 39-43, 1 1 2 , 1 1 4, 1 1 6
H e a d , s e e N e c k
H e a r t b e a t , m y o g e n i c , 5-6
b e g i n s i n , c a t s h a r k , 8
c h i c k , 28
m a n , 1 01
e f f e c t s o f o x y g e n o n , 57
H u m a n e m b r y o s a n d f e t u s e s , g e n e r a l ,
34, 36, 42, 54-86, 87, 93-97
108-109
a g e , d e t e r m i n a t i o n o f , 62, 77
jVi w e e k s , 55-56, 62-64, 67, 73,
78- 80, 1 08- 1 1 0
8 w e e k s , 55-56, 64, 67, 78, 79,
83, 95, 1 08- 1 1 0
&l w e e k s , 56, 64-68, 95, 108-
1 1 0
9V2 w e e k s , 65, 66-68, 78, 79,
1 0 8 - n o , 1 1 1 , 1 1 7
10V2 w e e k s , 68, 78, 80, 82, 83,
85, 1 08- 1 1 0
n w e e k s , 68-69, 78, 80, 82, 83-
84, 85, 1 04, 1 08-1 1 0
1 2 w e e k s , 70-71, 74, 80, 82-83,
86, 1 08- 1 1 0, 1 1 2 , 1 1 6
1 3 w e e k s , 71 -72, 74, 80, 82-83,
85, 1 08- 1 1 0, 1 1 2
1 4 w e e k s , 71 -73, 78, 80, 82, 85,
1 08-1 1 0, 1 1 8
1 5 w e e k s , 73-74, 78, 1 1 0, 1 1 1 ,
1 1 8
1 6 w e e k s , 74, 1 1 0
1 7-1 8 w e e k s , 74, 80, 81-82, 83,
1 1 0
18^2-20 w e e k s , 74-75, 80, 83,
85, 1 1 0
20V2-23V2 w e e k s , 58, 74-75
8 o

81 , 82, 83, 1 1 0, m
25 w e e k s , 76-77, 81-82, 1 1 1
27 w e e k s , 77, 82, 85
o l d e r t h a n 27 w e e k s , 77, 81 , 83,
85
i n j u r y t o , 59, 60, 61
n u m b e r o b s e r v e d , 60-61
r o t a t i o n , e f f e c t s o f , 66-67
s e q u e n c e o f a c t i v i t i e s i n , 62-86
H y p o x i a , s e e A n o x i a
I n d i v i d u a t i o n o f r e f l e x e s , 1 2, 21 , 52
I n h i b i t i o n , o f C h E . , 1 3
n e r v o u s , 99
I n t e g r a t i o n , p r i m a r y , 21 , 52
s e c o n d a r y , 53
I n t e r c h a n g e , 0 2 - C 0 2 , 53, 1 01 - 1 04
I n t r a - u t e r i n e a c t i v i t y , 106-107, 1 1 0,
1 1 2 - 1 1 5
p r e s s u r e , 59
K i l l i f i s h , 9-13
L i m b , fl ipping, 3 1 , 36, 46-47, 48, 73,
97 99
m o v e m e n t s , u n s p e c i f i e d , 20, 27, 41 ,
43 5
6
73 97-98> 100
l o w e r e x t r e m i t y , 25, 26, 29, 30,
3
1
3
2
35> 37 38, 39 4 44
49 5> 54 5
6
6
3"77 85-86
u p p e r e x t r e m i t y , 25, 26, 29, 30,
3
1
32, 33 35 37 3
5
> 39
40, 42, 43, 44, 48-49, 50,
56, 63-77, 79, 83-85, 91 ,
98
L i z a r d s , 25
L o g g e r h e a d t u r t l e , s e e Caretta caretta
L u n g s , 73, 81-82, 1 1 1 , 1 1 3 - 1 1 4
M a m m a l s , 33-120
M a r s u p i a l s , 34, 1 1 6- 1 1 7
M a n , a d u l t , 92, 99
e m b r y o n i c , s e e h u m a n e m b r y o s , e t c .
f e t a l , s e e h u m a n e m b r y o s , e t c .
M o r o r e f l e x , 76
M o r p h i n e , 58, 64
M o u t h , m o v e m e n t s o f , 26, 37, 3S,
39 44 5o-5i 71 7
2
> 74 75
76, 77, 82-83
M o v e m e n t s , s e e u n d e r A c t i v i t y , E y e ,
L i m b , M o u t h , e t c .
M u s c l e , a c t i v i t y , s e e u n d e r A c t i v i t y
d e v e l o p m e n t , 88-89
M y o g e n i c o r " M y o g e n i c , " s e e u n d e r
A c t i v i t y
N e c k , b e n d i n g , 28-29, 40
e x t e n s i o n , 26, 28, 35, 3S, 39, 40,
49 5 7 7 i
[140]
Subject Index
flexion, contralateral, 1 5 , 16, 25,
35 37 5
6
3
6
4"
6
5
6
7 7
1

79
8 o
> 9
s
ipsilateral, 1 5, 16, 50, 64, 67, 80
lateral, 8, 22, 26, 31 , 37, 42-43,
44, 48, 56
side-to-side, 8, 20, 31 , 35, 40, 44
ventral, 30, 31 , 35, 37, 40, 71
rotation, 28, 29, 35, 40, 44, 49, 50,
51 , 70, 72, 80
Nerve(s), cervical, 1 9, 33, 45, 95,
96, 97, 9S
collaterals, 1 7, 33, 96
commissural, 95-96, 98
cranial, V, 1 7, 18, 50-51, 63, 68,
69, 70, 78, 79, 90, 91-97, 98,
1 05
V
a
, 51 , 68, 69, 78, 91-97, 98
V 50-52, 63, 68, 69, 70, 78,
79 9
l
~97> 9
8
V
8
, 51 , 63, 68, 69, 70, 78, 79,
91-97, 98
VI I , 1 7, 18
X, 1 7, 18
XI , 33, 96
motor, see Neurons, efferent
sensory, see Neurons, afferent
tracts, reticulo-spinal, 93, 97
spinal V, 1 7, 1 8, 91-97, 98
Neural mechanism, 6-7, 1 7, 18, 20,
2 1 , 104, 1 05
Neurogenic movements, see under
Activity
Neuromuscular mechanism, 3, 1 3, 1 4,
1 7, 18, 1 9, 20, 52, 73, 87-90,
91-97, 98, 1 02, 107, 109
Neuron, afferent, 7, 9, 1 7, 32, 33,
89-97, 105
effector, 7, 1 0, 1 1 , 90
efferent, 6-7, 1 0, 1 1 , 1 9, 33, 89,
90
intercalated, see internuncial
internuncial, 7, 90
motor, see efferent
receptor, 7, 90
sensory, see afferent
Nitrous oxide, 58
Novocaine, 58-59
Opsanus, 9-12, 35
Opossum, 34
Oxygen, effects of, 57, 80
requirements, 102-105
want, see Anoxia
Pattern, partial, 1 2, 21 , 48, 52, 7 1 , 7 2
total, 1 2, 19-20, 21 , 24, 26, 42,
43 45 47 4
8
5
1
"53
6 8
6
9
7
l
7
2
> 73
Pelvis, see Rump
Phonation, 44, 73, 75
Picrotoxin, 58
Pig, 1 1 2 , 1 1 6- 1 1 7, 1 1 9
Pigeon, 28, 30-31
Placenta, 47, 55, 56, 57, 65, 66, 74,
75, 80, 102-103, 106, 107-108,
109, 1 1 1 , 1 1 5 , 1 1 7 - 1 1 8
Placental barriers, 102-103
Proprioceptive, see under Reflex
Protoplasmic bridges, 9
Pseudacris, 22
Quickening, 106
Rabbit, 37-38, 41 , 92, 1 1 3 - 1 1 4
Rana, 22-24, 84, 107
Rat, general, 34-37, 38, 39, 41 , 84,
88, 89, 90, 100-101, 105, 107-
108, 1 1 7 , 1 1 8
sequence of activities, 35-37
Recording, dictation, 60
motion pictures, 55, 60
Reflex (es), arc, 7, 32, 45, 90, 97, 98,
99
1 0
4
exteroceptive, 6, 7, 9, 1 0, 1 1 , 1 3 ,
14-24, 29, 31 , 36, 37, 38,
39-40 41-42, 63-77, 79"
8 6
>
s
9
90, 91-99, l oo-i oi , 105, 108-
109, 120
human fetal, cremasteric, 77
earliest, 55-57, 62-64, 79
eye, 71 , 76, 83
eyelid, 68, 69, 71 , 72, 76, 78,
82, 83
face, 68, 69, 70, 71 , 72, 74, 75,
76, 77, 82-83
finger, 68, 7 1 , 72, 73, 75, 83-85
grasp, 72, 73, 75, 77, 84-85
intra-uterine, 106-107
lower extremity, 54, 56, 63-77,
85-86
Moro, 76
mouth, 71 , 72, 74, 75, 76, 77,
82-83
neck, 56, 63, 64-65, 70, 7 1 , 72,
79
8 0
palmar, 68, 78, 83-85
phonation, 73, 75
plantar, 54, 68-69, 70, 71 , 85-86
[ HI ]
Subject Index
REFLEXOGENOUS AREAS OF SKIN, 56,
63, 64, 65-68, 69, 70, 72.
73 77-79
o
RESPIRATORY, 74, 75-76, 80-82
RIGHTING, 66-67
RUMP, 64-67, 79, 85
STRETCH, 67, 79, 97-98, 99
SUCKLING, 73, 75, 77, 83
SWALLOWING, 71 , 72, 83
TRUNK, 56, 60-71, 75-76, 79, 80,
83, 96, 98
UPPER EXTREMITY, 56, 63-77, 7^
79 83-85, 98
VESTIBULAR, 66-67, 76
INTEROCEPTIVE, 100
PROPRIOCEPTIVE, 20, 36, 67, 79, 90,
97-98, 99, 100
SIMPLE, 36, 43, 45, 47, 48, 52, 53,
97, 99-100
SPECIFIC, SEE UNDER PATTERN, PARTIAL
STRETCH, 36, 67, 79, 97-98, 99
TENDON, 73, 76, 1 1 0
UNITARY, SEE SIMPLE
REFLEXOGENIC AREAS, SEE SENSORY AREAS
OF SKIN
MOVEMENTS, SEE REFLEXES, EXTERO-
CEPTIVE
REPTILES, 25-27
RESPIRATION, 36, 58, 73, 74, 75-76,
80-82, 1 02, 1 1 0 , 1 1 2 - 1 1 5
RESPIRATORY MOVEMENTS, 30, 49, 74,
75-76, 80-82, 1 1 2 - 1 1 4
RESPONSES, SEE UNDER ACTIVITY, AND
REFLEX
RESUSCITATION, 58, 74, 75, 82
RHYTHMIC ACTIVITY, 5, 6, 7-8, 9, 10,
1 2, 27, 49, 1 1 3 - 1 1 4
RIGHTING MOVEMENTS 23, 26, 38, 39,
66-67
ROHON-BEARD CELLS, 1 7, 18
RUMP ROTATION, 35, 38, 39, 64-67, 79,
85
SCYJJIORHINUS, 7-10
SENSORY AREAS OF SKIN, 1 1 , 1 4, 1 5, 17-
18, 23, 25, 26, 3 1 , 32, 37, 42,
50, 51 , 56, 63, 64, 65, 66,
67, 68, 69, 70, 72, 73, 77-79
SHEEP, GENERAL, 46-47, 48-51, 66-67,
74 93 99
1 0
4 " 3
SEQUENCE OF ACTIVITIES, 48-51
SNAKES, 25
SNOUT TAPPING, 46
SOLUTIONS, 56, 59
SPINAL ACCESSORY NERVE, SEE NERVE,
CRANIAL, XI
SPINAL CORD, 9, 1 7, 18, 20, 23, 24, 33
47, 89, 91 , 92-97, 104-105
SPINAL GANGLIA, 89-90, 91
SPONTANEOUS MOVEMENTS, SEE UNDER
ACTIVITY
STEREOTYPED RESPONSES, 1 5, 64, 65,
7 1 , 80
STIMULATION, ELECTRICAL, 6, 1 4, 30, 31 ,
3
2
37 39"4 4
2
45"4
6
49*5
61-62, 63, 89
ENDOGENOUS, 1 0- 1 1 , 35-36, 49
EXTEROCEPTIVE, 1 1 , 1 4-1 5, 20, 22, 25,
2
9 3 3
6
37 39"4 4
1
6 l
*
62, 63, 66, 1 00-1 01 , 108-109,
1 20
MECHANICAL, 6, 1 4, 3 1 , 32, 33, 36,
37 39-4 43-45 47 5
6
6
4
79, 89, 1 01 , 108
METHODS, 1 4- 1 5, 22-23, 3"*3
1
> 35"
36, 39-40, 42, 47, 56, 61-62,
6
3 97
PRESSURE, 22-23, 3 39 4
2
> 45
56, 61-62
PROPRIOCEPTIVE, 20, 36, 67, 97
ROTATION, 66-67
THRESHOLD, 34, 65, 74, 109
STIMULATORS, GENERAL, 1 4, 39-40, 61-
62
GLASS RODS, 50-51, 61 , 75
HAIRS, GENERAL, 1 5 , 31 , 35, 39, 67-
68, 108
CALIBRATED, 61-63, "5 7
2
75 79'
80, 109
OTHER, 35, 36, 39-40, 43, 56
TACTUAL, SEE HAIRS
USE OF, 39-40, 42, 61-63, 65, 67-68,
79, 80
STRETCH REFLEXES, SEE UNDER REFLEX
SUCKLING RESPONSES, 51 , 73, 75, 77, 83
SWALLOWING, 29, 30, 51 , 7 1 , 72, 83,
1 1 1 - 1 1 2
TAUTOGOLABRUS, 9, 1 0
TEMPERATURE CONTROL, 59
TOADFISH, SEE OPSANUS
TRACTS, SEE UNDER NERVE
TRANSAMINATION, 28
TRAUMA, 59-60, 109
TRIGEMINAL NERVE, SEE NERVE, CRANIAL,
V, ETC.
TROUT, 1 0
TRUNK, BENDING, 29
EXTENSION, 29, 35, 50, 68-71, 75-
76, 80
[142]
Subject Index
flexion, contralateral, 1 5 , 26, 1 7,
2
5>
2 ( 5
> 35 37 47 5
6
63-71,
79, 80, 96, 98, 100-101
ipsilateral, 1 5, 16, 26, 47, 50, 56,
64, 67, 80, 100-101
lateral, 8, 1 0, 23, 26, 30, 31 , 32,
35 37 3
8
> 39 4' 4>43
44, 48-49, 56, 60-68, 69, 76
side-to-side, 7, 8, 9, 1 0, 1 9, 33,
35 4 44 49
6
5 "
6 6
ventral, 32, 35, 37, 38, 39, 44,
56, 76
unspecified, 41 , 49
Turtles, 25-27
Urinary system, m
Urodela, 14-25
Vagus nerve, see Nerve, cranial, X
Vestibular responses, 66-67, 76
Viability, 76-77, 82, 1 1 1
Yolk sac, contractions, see under Chick.
[143]

The Prenatal Origin of Behavior by Davenport Hooker, PH.D., Sc.D. - University of Kansas Press, Lawrence, Kansas, 1952

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Prenatal Origin of Behavior by Davenport Hooker, PH.D., Sc.D. - University of Kansas Press, Lawrence, Kansas, 1952

Uploaded by

Copyright:

Available Formats

Tne Prenatal Origin

You might also like